JP7220822B1 - Program, information processing device, and information processing method - Google Patents

Abstract

A program, an information processing apparatus, and an information processing method are provided, which enable customers to be suitably provided with health-related information. SOLUTION: In a health prediction system, an integrated health prediction program includes a process of acquiring first purchase data representing a purchase history of a target customer, and from the first purchase data, first to N types (N is 2 or more). Prediction processing for predicting the degree of multiple types of health symptoms using a health prediction model for multiple types of health symptoms (integers) and processing for providing target customers with health information indicating the degree of the predicted multiple types of health symptoms. Let the information processing device execute. [Selection drawing] Fig. 7

Description

Some aspects of the present invention relate to programs, information processing apparatuses, and information processing methods.

In recent years, living environments around the world are undergoing drastic changes, such as population decline, economic depression, the occurrence of natural disasters, and the spread of infectious diseases. On the other hand, opportunities for utilizing artificial intelligence (AI) are increasing, and the development and spread of fifth-generation mobile communication systems are progressing. As our living environment continues to change, the infrastructure that supports society has begun to shift to an industrial framework that utilizes technology and is based on data, and social activity and behavioral information is beginning to be visualized.

As interest in health increases, it is required to adapt health management to social trends. Until now, the focus has been on "passive health care," in which people are treated only after the disease has begun to worsen. However, in the future, healthy people will obtain their own data at home, collect that data, and digitize the health status of each person. It is thought that “preventive medicine” will also be implemented. In the future, "predictive medicine" that predicts future diseases and calls attention to health can also be envisioned.

Patent Literature 1 discloses a method of constructing an estimation model for estimating the target user's health condition based on data describing the target user's purchase history. This patent document is hereby incorporated by reference in its entirety.

Patent No. 6916367

There are various types of risks in health conditions, but the method described in Patent Document 1 does not consider the customer's grasp of various health risks.

Some aspects of the present invention have been made in view of any of the problems described above, and provide a program, an information processing apparatus, and an information processing method that enable customers to preferably receive health-related information. to provide.

A program according to an exemplary embodiment of the present invention includes a process of acquiring first purchase data representing a purchase history of a target customer; prediction processing for predicting the degree of the plurality of types of health symptoms using a health prediction model for the plurality of types of health symptoms (integers); and providing health information indicating the predicted degree of the plurality of types of health symptoms to the target customer. A program for causing an information processing apparatus to execute a process to be provided, wherein the health prediction model for a k-th type (1≦k≦N) health symptom is obtained from the first purchase data concerning the product category, the k-th type It is a program that calculates a numerical value that indicates the degree of health symptoms of a person.

An information processing apparatus according to an exemplary embodiment of the present invention includes an input unit that acquires first purchase data representing a purchase history of a target customer; a prediction unit that predicts the degree of the plurality of types of health symptoms using a health prediction model for the plurality of types of health symptoms (integers of 2 or more); and an output unit to be provided to a target customer, wherein the health prediction model regarding a kth type (1≤k≤N) of health symptoms is obtained from the first purchase data regarding the product category, the kth Calculate a number that indicates the severity of the health symptoms of the species.

An information processing method according to an exemplary embodiment of the present invention includes a process of acquiring first purchase data representing a purchase history of a target customer; a process for predicting the degree of the plurality of types of health symptoms using a health prediction model for the plurality of types of health symptoms, and sending health information indicating the predicted degree of the plurality of types of health symptoms to the target customer. , wherein the information processing device performs a health prediction model related to the kth type (1 ≤ k ≤ N) of health symptoms, wherein the health prediction model related to the product category is the first A numerical value indicating the degree of the k-th type health symptom is calculated from the purchase data.

In the present invention, "part", "means", "apparatus" and "system" do not simply mean physical means. This includes the case where the function of "" is realized by software. In addition, even if the function of one "part", "means", "apparatus", or "system" is realized by two or more physical means or devices, two or more "part", "means", "device", or "system" The functions of "apparatus" and "system" may be realized by one physical means or device.

According to the exemplary embodiments of the present invention, it is possible to provide a program, an information processing device, and an information processing method that enable customers to be suitably provided with health-related information.

It is a figure which shows typically the flow of the data for performing a customer's health prediction using the health prediction system which concerns on embodiment. It is a figure which shows the specific example of functional structure of the integrated health prediction program which concerns on embodiment. It is a figure which shows the specific example of the purchase data which concerns on embodiment. It is a figure which shows the specific example of the purchase data which concerns on embodiment. FIG. 10 is a diagram showing a specific example of a table for realizing proposals to customers according to output values of prediction models; FIG. 10 is a diagram showing a specific example of a table for realizing proposals to customers according to output values of prediction models; It is a figure which shows the flow of a process of the integrated health prediction program which concerns on embodiment. It is a figure which shows the specific example of the hardware constitutions of the information processing apparatus by which the integrated health prediction program which concerns on embodiment is executed. 1 is a diagram showing a specific example of functional configuration of a learning system including a model learning program according to an embodiment; FIG. It is a figure which shows the example of a goods category. FIG. 3 is a diagram showing a specific example of the relationship between menopausal symptoms found from items purchased by H-tier subjects regarding menopausal symptoms and health issues. FIG. 10 is a diagram showing a specific example of a product category for women with menopausal symptoms belonging to the H tier. It is a figure which shows the flow of a process of the model learning program which concerns on embodiment. It is a figure which shows the specific example of the hardware constitutions of the information processing apparatus with which the model learning program which concerns on embodiment is performed. FIG. 15 is a block diagram showing an example of the configuration of a communication system according to one embodiment. FIG. 16 is a block diagram showing an example of the hardware configuration of the server device 1510 shown in FIG. 15. As shown in FIG. FIG. 17 is a flow diagram illustrating an example of operations performed by the communication system 1500 shown in FIG. FIG. 18 is a diagram showing an example of the Athens Insomnia Scale. FIG. 19A is a diagram showing an example of the Pittsburgh Sleep Questionnaire. FIG. 19B is a diagram showing an example of the Pittsburgh Sleep Questionnaire. FIG. 19C is a diagram showing an example of the Pittsburgh Sleep Questionnaire. FIG. 20A is a diagram showing an example of a three-dimensional sleep scale. FIG. 20B is a diagram showing an example of a three-dimensional sleep scale. FIG. 21 is a diagram showing an example of an insomnia severity questionnaire. FIG. 22A shows an example of the Munich Chronotype Questionnaire. FIG. 22B is a diagram showing an example of a Munich chronotype questionnaire. FIG. 23 is a diagram showing an example of sleep category data used in communication system 1500 shown in FIG. FIG. 24 is a diagram showing another example of sleep category data used in communication system 1500 shown in FIG. FIG. 25 is a diagram conceptually showing an example of a search table used in communication system 1500 shown in FIG. FIG. 26 is a flow diagram illustrating another example of operations performed by the communication system 1500 shown in FIG. FIG. 27 is a diagram showing an example of the Simplified Menopause Index (SMI). FIG. 28 conceptually shows another example of a search table used in communication system 1500 shown in FIG.

The various systems, methods and devices described herein should not be construed as limited in any way. In practice, the present disclosure is directed to each of the various disclosed embodiments, combinations of these various embodiments with each other, and combinations of portions of these various embodiments with each other. All novel features and aspects are addressed. The various systems, methods, and apparatus described herein are not limited to any particular aspect, particular feature, or combination of such particular aspects and particular features, The articles and methods described herein do not claim that one or more particular advantages exist or that any problem is solved. Moreover, various features or aspects of the various embodiments described herein, or portions of such features or aspects, may be used in combination with each other.

Although the operations of some of the various methods disclosed herein have been described in a particular order for convenience, the description in such a manner is such that the particular order follows a particular order. It should be understood to include rearranging the order of the above operations unless otherwise required by the text. For example, operations described sequentially are in some cases permuted or performed concurrently. Furthermore, for purposes of brevity, the accompanying drawings do not illustrate the various ways in which the various features and methods described herein can be used with other features and methods.

Any theories of operation, scientific principles or other theoretical statements presented herein in connection with the disclosed devices or methods are provided for better understanding and to limit the technical scope. not intended to be. The devices and methods in the appended claims are not limited to devices and methods that operate according to methods described by such theory of operation.

Any of the various methods disclosed herein can be implemented using a plurality of computer-executable instructions stored on one or more computer-readable media and executed on a computer. can be The one or more media are non-transitory computer-readable storage media, such as, for example, at least one optical media disc, volatile memory components, or non-volatile memory components. obtain. Here, the plurality of volatile memory components includes, for example, DRAM or SRAM. The plurality of non-volatile memory components also includes, for example, hard drives and solid state drives (SSDs). Further, the computer includes any computer available on the market, including, for example, smartphones and other mobile devices having computing hardware.

Any of such computer-executable instructions for implementing the techniques disclosed herein may be generated and used during implementation of the various embodiments disclosed herein. Any data may be stored on one or more computer-readable media (eg, non-transitory computer-readable storage media). Such computer-executable instructions may, for example, be part of a separate software application, or may be accessed or downloaded via a web browser or other software application (such as a remote computing application). can be part of a software application that is Such software may be implemented, for example, in a network environment, either on a single local computer (eg, as a process running on any suitable computer available on the market) or using one or more network computers. (eg, the Internet, a wide area network, a local area network, a client-server network (such as a cloud computing network), or other such network).

For clarity, only certain selected aspects of various software-based implementations are described. Other details that are well known in the art are omitted. For example, the technology disclosed herein is not limited to any particular computer language or program. For example, the techniques disclosed herein may be implemented by software written in C, C++, Java, or any other suitable programming language. Similarly, the technology disclosed herein is not limited to any particular computer or type of hardware. Certain details of suitable computers and hardware are well known and need not be described at length here.

Further, various such software-based embodiments (e.g., including computer-executable instructions for causing a computer to perform any of the various methods disclosed herein) can be uploaded, downloaded, or accessed remotely by any suitable communication means. Such suitable means of communication include, for example, the Internet, World Wide Web, intranets, software applications, cables (including fiber optic cables), magnetic communications, electromagnetic communications (including RF communications, microwave communications, infrared communications), Including electronic communication or other such means of communication.

1. Overview FIGS. 1 to 14 are diagrams for explaining an embodiment. Hereinafter, broadly speaking, a health prediction system that predicts the customer's health condition (also referred to as health symptoms) and a learning system that generates a prediction model for predicting the customer's health condition, which is used in the health prediction system Separately, the overview, functional configuration, processing flow, etc. will be described in order.

2 Health prediction system
2.1 Overview of Processing of Health Prediction System First, an overview of the processing of the health prediction system 1 according to the embodiment will be roughly described with reference to FIG. FIG. 1 schematically shows the flow of data for predicting the health of a customer using the health prediction system 1 according to the embodiment.

The health prediction system 1 is for estimating an individual's health condition from the purchase behavior of the individual. By using this health prediction system 1, it is possible to estimate a customer's individual health condition based on purchase data, which is the purchase behavior of the customer, without data relating to the customer's health such as examination data. The estimated health condition can be used, for example, to improve the customer's health.

Here, the integrated health prediction program 100 for predicting an individual's health condition from an individual's purchasing behavior includes a plurality of types of learned prediction models for predicting the health condition from an individual's purchasing behavior. In the example of FIG. 1, the integrated health prediction program 100 is configured by integrating six prediction models, but the number is not limited to six, and may be five or less, or seven or more. may

The integrated health prediction program 100 according to the present embodiment includes a female health prediction model 110a that predicts women's health, a sleep state prediction model 110b that predicts sleep state, a lifestyle disease prediction model 110c that predicts lifestyle diseases, a water Six prediction models (hereinafter collectively referred to as "prediction models 110"): a moisture state prediction model 110d that predicts the state, an immune state prediction model 110e that predicts the immune state, and a nutritional state prediction model 110f that predicts the nutritional state. including. Each prediction model included in the integrated health prediction program 100 is based on the purchase data 150 indicating the customer's purchase history, women's health condition (such as SMI related to menopausal symptoms)/sleep condition/lifestyle disease/hydration condition (hydration condition). /immune status/nutritional status (value indicating the possibility and/or degree of suffering from a specific disease) is calculated as the customer's health prediction value 160. FIG. In the example of FIG. 1, the six health status values are shown in %. Depending on the type of disease, only men, only women, or both men and women may be targeted. These predictive models 110 can be described as functions. It is constructed by machine learning using teacher data including product purchase history and various health conditions. A method for constructing the prediction model will be described later.

More specifically, integrated health prediction program 100 receives customer purchase data 150 . The purchase data 150 is data related to the purchase history of a customer whose health is to be predicted, and is data corresponding to explanatory variables of teacher data when constructing each prediction model 110 included in the integrated health prediction program 100 . The integrated health prediction program 100 inputs purchase data 150 to each prediction model 110 in parallel to calculate a customer's health prediction value 160 . In this embodiment, only the purchase data 150 is input to the prediction model 110, but customer attribute information such as the customer's gender and age may also be input.

In addition, the integrated health prediction program 100 may make various proposals regarding health improvement and the like to the customer according to the calculated health prediction value 160 . More specifically, for example, it is possible to point out a category with a high health prediction value to the customer and propose measures that can be taken while the customer is not aware of (latent) symptoms or symptoms that are manifesting are mild. Conceivable. The proposed method will be described later with reference to FIG. 4 and the like.

2.2 Configuration of Integrated Health Prediction Program The functional configuration of the integrated health prediction program 100 will be described below with reference to FIG. FIG. 2 is a diagram showing the functional configuration of the integrated health prediction program 100. As shown in FIG. Integrated health prediction program 100 includes input section 120 , product category determination section 125 , prediction model 110 , output section 130 and proposal section 135 . The integrated health prediction program 100 may be implemented as a single program, or may be implemented as a plurality of programs that can operate in cooperation by inputting and outputting data.

The input unit 120 receives (acquires) input of purchase data 150 . The input unit 120 may read the purchase data 150 from a storage medium such as a HDD (Hard Disk Drive) or an SSD (Solid State Drive) built into the information processing device (computer) on which the integrated health prediction program 100 is executed. Alternatively, purchase data 150 may be input from an external information processing device or the like connected via a LAN (Local Area Network), the Internet, or the like.

Here, a specific example of the purchase data 150 that the input unit 120 receives will be described with reference to FIG. FIG. 3 is a diagram showing a specific example of purchase data 150. As shown in FIG. Here, as the purchase data 150, panel data provided by a research company, in which customer's purchase histories are accumulated, can be adopted. The panel data is data acquired by scanning the product by the customer who is the purchaser when the product is purchased. When a customer scans a product using their own smartphone or a lent bar code reader, the product is associated with preset customer attributes (employee/student/housewife, etc.), purchaser information, etc. are accumulated. As a result, it is possible to grasp what kind of person purchased what, how much, and at what kind of store.

The example of purchase data 150 in FIG. 3 indicates the purchase history of foodstuffs and/or daily necessities purchased by a customer over a period of time (eg, one year). Each product purchased by a customer is classified into product categories in JICFS (JAN Item Code File Service) classification. In other words, at least one product belongs to each product category. In the purchase data 150, about 500 sub-categories (e.g., soft drinks, Mouthwash, etc.) are associated with actual data 150b, which is numerical data such as purchase price, purchase quantity, and number of purchases during the period.

In the present embodiment, the purchase data 150 includes the purchase history of food and/or daily necessities, but is not limited to these as long as the relationship with the customer's health condition can be estimated. . For example, the purchase data 150 may include purchase histories of other products such as medicines and various services such as massage and acupuncture and moxibustion treatment. The purchase data 150 may be identified by payment information managed by the retail store, or may be identified by using credit card payment information, payment information to a financial institution, or the like.

When the product category determination unit 125 receives the purchase data 150 from the input unit 120, the product category determination unit 125 determines the classification code (JICFS classification) and product category to be input to each prediction model 110 from the purchase data 150 to be input to each prediction model 110, Actual data 150b (only some of them may be used), which is numerical data such as purchase amount, purchase quantity, and number of purchases related to the obtained classification code and product category, is input to each prediction model 110 in parallel. do.

The classification code (product category) of the purchase data 150 to be input to the prediction model 110 can be determined for each prediction model 110 . The method of determining the product category to be input to the prediction model 110 will be described later in 3.4 regarding the creation of the prediction model 110 below.

It is conceivable that the purchase data 150 input from the input unit 120 may be directly input to the prediction model 110 without being narrowed down by product category. In that case, the product category determination unit 125 is unnecessary.

The predictive model 110 receives the input of performance data 150b, which is numerical data such as the purchase amount, the purchase quantity, and the number of purchases related to the classification code and product category selected by the product category determination unit 125. symptoms), sleep status (disturbed sleep), lifestyle-related diseases (lack of physical activity), hydration status (level of dehydration/hydration status), immune status (level of decreased immunity), and nutritional status (level of nutritional deficiencies) Calculate the predicted value. In the example of FIG. 1, for the customer with the user ID 001, each prediction model 110 is 90% for menopausal symptoms, 21% for disturbed sleep, 40% for lack of physical activity, 76% for lack of water, 72% for immunosuppression, A value of 11% undernutrition is calculated.

As described above, the prediction model 110 uses the purchase data 150, which is the purchase history of each customer, and the teacher data that associates the health conditions of those customers, with the former as an explanatory variable and the latter as an objective variable. Machine learning It can be constructed by A method for constructing the prediction model 110 will be described later in 3 below.

The output unit 130 outputs the value calculated by each prediction model 110 as a health prediction value 160 for the customer to the proposal unit 135, a built-in storage medium, other information processing devices connected via a network, etc. Output. Alternatively, the output unit 130 may display the calculated health prediction value 160 on the display device. In the example of FIG. 1, the output unit 140 sequentially arranges the values of the customer ID, menopause, sleep disturbance, etc. in the row direction, and repeats this for each ID of the customer data, thereby predicting the health of a plurality of customers. It outputs a value of 160. Note that the predicted health value 160 may include not only the actual health condition of the customer but also the health condition that can be expected to occur in the future (potential health condition).

The proposal unit 135 proposes foods, goods, services, and the like according to the predicted health prediction value 160 for each customer. Various proposal methods are conceivable. For example, the proposal may be displayed on various terminals used by retail store clerks, the content of the proposal may be sent directly to the customer via a messenger service, etc., or the proposal may be posted on a web page for the customer. It is also conceivable to display it on the

2.3 Specific Examples and Processing of Various Proposals to Customers Hereinafter, various proposals made by the integrated health prediction program 100 to customers will be described with reference to FIGS. 4 to 6. FIG.

FIG. 4 is a diagram schematically showing foods/products/services etc. proposed by the proposal unit 135 according to the customer's health prediction value 160. As shown in FIG. For example, for a woman (customer ID=001) whose menopausal symptoms are predicted to be 90%, the proposal unit 135 proposes products sold at the drug store 41 that improve menopausal symptoms. For a man (customer ID=002) whose malnutrition level is predicted to be 93%, the proposal unit 135 proposes highly functional nutritional foods sold at the drugstore 41 . As another example, the suggestion unit 135 suggests that a woman (customer ID=003) whose predicted values for menopause and sleep disturbance are 79% and 89%, respectively, consult a medical institution 43 . In addition, the proposal unit 135 proposes that a customer with a relatively high degree of lack of physical activity go to the gym 45, and the insurance company 47 proposes that a customer with a relatively low degree of lack of physical activity go to the gym 45. , proposes medical insurance products for lifestyle-related diseases.

In this way, the proposal unit 135 of the integrated health prediction program 100 obtains prediction results from each of the plurality of prediction models 110 from the purchase data 150, thereby proposing food/articles/services suitable for the customer. can. To implement such suggestions, a list of foods, goods, and/or services associated with each health condition, which is the output of the predictive model, may be prepared in advance.

FIG. 5 shows a table 50 and referenced databases 52, 54 and 56 for implementing suggestions regarding menopausal symptoms.

In the table 50, the output range of the menopausal prediction model (women's health prediction model 110a in FIG. 2) is less than 40% class 50a, 40% or more and less than 80% class 50b, and 80% or more class 50c. There are two classes. In class 50a, a product database 52 for preventing menopausal symptoms in a drug store is referenced. In the product database 52, products X1 and X2 of product category P1 and product X3 of product category P2 are described. The proposal unit 135 can refer to the product database 52 and propose some or all of the categories and/or product names to the corresponding customer. When the output value of the prediction model corresponds to the class 50b, the proposal unit 135 refers to the product database 54 for improving menopausal symptoms and proposes products Y1 and Y2 of product category Q1. If the output value of the prediction model corresponds to 50c, the proposal unit 135 refers to the menopausal outpatient medical institution database 56 using information on the current location and/or the customer's residence registered in advance, and proposes a medical institution. In other words, this corresponds to referral to the customer to a medical institution that performs preventive care and/or treatment according to the customer's health prediction result.

Another example is described with reference to FIG. FIG. 6 shows a table 60 for realizing a proposal according to the output value of the lifestyle-related disease prediction model 110c.

In the table 60, the output values of the lifestyle-related disease prediction model 110c, in other words, the class 60a where the range of lack of physical activity is less than 20%, the class 60b where the range is 20% or more and less than 80%, and the class where the range is 80% or more There are three classes of 60c. In class 60a, physical activity is considered to be substantial, and in order to encourage continued physical activity, nutritional guidance (1) and a monthly fee plan for the E1 sports gym are proposed. Nutritional guidance (1) is a dietary suggestion, for example, to take nutritional components of less than 10 grams of fat and 30 grams or more of protein per meal. In class 60b, it is considered that the physical activity is sufficient, and an E2 fitness club membership fee plan is proposed for the purpose of providing nutritional guidance (2) and regular physical activity. Nutritional guidance (2) is, for example, a diet suggestion for ingesting nutritional components of less than 20 grams of fat and 20 grams or more of protein per meal. In class 60c, physical activity is insufficient, and based on the possibility of contracting lifestyle-related diseases in the future, nutritional guidance (3) and medical insurance products proposed by E3 Life Insurance Company are proposed. Nutritional guidance (3) is, for example, a diet suggestion for taking nutritional components of less than 20 grams of fat, 20 grams or more of protein, and less than 20 grams of carbohydrate per meal. Nutrition coaching, gyms, fitness clubs and insurance product offerings are categories of product or service offerings. With regard to the output value of the lifestyle-related disease prediction model 110c as well, the proposal unit 135 can refer to the table 60 to make a proposal to the customer according to the output value.

In addition to the examples in FIGS. 5 and 6, the proposal unit 135 can provide each customer with a similar table and a database as necessary for each of sleep state, immune state, hydration state, and nutritional state. A proposal can be made to the customer according to the health prediction value 160 that is the output value of the prediction model 110 .

Note that the integrated health prediction program 100 itself does not necessarily need to include the proposal unit 135 . For example, the output unit 130 of the integrated health prediction program 100 transmits the health prediction value 160 to a business operator other than the business operator that operates the integrated health prediction program 100 with the permission of the customer, and It is also conceivable that the program of the business implements the program related to the proposal unit 135 and makes the above proposal to the customer. For example, the integrated health prediction program 100 may transmit the health prediction value 160, which is the output value of the prediction model 110, to an insurance company so that the insurance company may propose an insurance product to the customer.

2.4 Processing Flow The processing flow of the integrated health prediction program 100 will be described below with reference to FIG. FIG. 7 is a flow chart showing the procedure of a customer's health prediction process executed by the integrated health prediction program 100. As shown in FIG.

The input unit 120 of the integrated health prediction program 100 acquires the purchase data 150 indicating the purchase history of the customer whose health condition is to be predicted (S701). For example, the input unit 120 reads the customer's membership card from a POS (Point of Sale) system when purchasing a product at a retail store, thereby constructing a database linking the customer with the purchase history. By extracting the purchase data 150 from such a database with the customer's consent, the input unit 120 can input the purchase data 150 to the integrated health prediction program 100 . As previously mentioned, such purchase data 150 includes, for example, JICFS classifications and product categories.

The product category determination unit 125 determines the JICFS classification (classification code) and product category to be input to each prediction model 110 from the purchase data 150 (S703). As described above, if the prediction model 110 does not specifically narrow down the product category used for prediction and uses the entire purchase data 150, the processing of S703 is unnecessary.

Next, each prediction model 110 predicts the health condition using the classification code determined by the product category determination unit 125 and the purchase data 150 of the product category (S705). The output unit 130 outputs the health prediction value 160 indicating the health condition predicted by the prediction model 110, and the proposal unit 135 provides the customer with a proposal based on the health prediction value 160 (S707). Specifically, it is conceivable that the health prediction value 160 or the proposal unit 135 proposes health improvement to the customer in the form of printing on a receipt, display on a display, notification to a smartphone, or the like.

2.5 Hardware Configuration A specific example of the hardware configuration of the information processing device 800 in which the integrated health prediction program 100 is executed will be described with reference to FIG. FIG. 8 is a hardware configuration diagram of an information processing device 800 on which the integrated health prediction program 100 is executed. The information processing device 800 uses the integrated health prediction program 100 to calculate prediction results regarding the client's menopausal symptoms, sleeping state, and the like. The information processing device 800 can be a commonly available computer system such as a desktop PC (Personal Computer), a notebook PC, a tablet PC, a server computer, or the like. The computer system may be installed at the office of the business operator that operates the health prediction system 1, or may be provided as a cloud service. In the latter case, the office may be provided with a PC or the like that can mutually communicate with the cloud service.

Information processing apparatus 800 includes control unit 810 , input interface (I/F) unit 820 , storage device 830 , and output I/F unit 840 .

The control unit 810 may include a CPU (Central Processing Unit, not shown), a ROM (Read Only Memory, not shown), a RAM (Random Access Memory), and the like. Control unit 810 can execute integrated health prediction program 100 stored in storage device 830 . Thereby, the information processing apparatus 800 can execute various processes related to health prediction described above in addition to the function as a general computer. Note that the arithmetic circuit included in the control unit 810 may not be a CPU, but may be realized by various processors such as an MPU or GPU, and may be realized by a plurality of processors instead of one processor.

The input I/F unit 820 receives purchase data 150 relating to a customer whose health condition is to be predicted. As described above, the input I/F unit 820 can receive the purchase data 150 from an information processing device such as an external server connected via a network such as the Internet or LAN. The input I/F unit 820 can be implemented by, for example, an Ethernet (registered trademark) communication terminal, a USB (registered trademark) terminal, or a communication circuit that performs communication in compliance with standards such as IEEE802.11, 4G, or 5G. can.

The storage device 830 is a storage medium that stores computer programs and data necessary for operating the information processing device 800 . The storage device 830 may be, for example, an HDD or an SSD, which is a semiconductor storage device. The storage device 830 may include a temporary storage element configured by RAM such as DRAM or SRAM, and may function as a work area for the control unit 810 . The storage device 830 stores the integrated health prediction program 100 and a proposal table 831 exemplified as the table 50 in FIG. 5 and the table 60 in FIG.

Note that the prediction model 110 may be incorporated as part of the integrated health prediction program 100 or may be provided as data separate from the integrated health prediction program 100 . Also, the prediction model 110 may be stored in the storage device 830 as a table or parameter group.

In addition, the proposal table 831 shows prediction results indicating the degree of health condition, for example, H class/M class/L class, 90%/50%/20%, and products/services to be proposed according to each degree. can be associated.

The output I/F unit 840 is a communication circuit and/or a communication terminal connected to various output devices provided outside the information processing device 800 . For example, the output I/F unit 840 may be a USB terminal that outputs print data indicating the content to be printed on the receipt printer P. FIG. The integrated health prediction program 100 can calculate a prediction result regarding the customer's health condition, and send the result or a proposal to the customer according to the result to the receipt printer P to print it on a receipt.

The output I/F unit 840 may be a video output terminal connected to the display D. The integrated health prediction program 100 calculates a prediction result regarding the customer's health condition, and displays on the display D the result or a proposal to the customer according to the result. For example, a pharmacist confirms the content displayed on the display D, and the pharmacist can propose one or more products according to the customer's symptoms.

As another form, the output I/F section 840 may be a communication terminal or a communication circuit capable of data communication by being connected to the communication network N or the like. When the output I/F section 840 is a communication terminal or communication circuit, the hardware of the input I/F section 820 and the output I/F section 840 may be the same. The output I/F unit 840 can transmit a prediction result regarding the customer's health condition or a proposal to the customer according to the result to the customer's smart phone M, for example, via a mobile phone line.

3 Learning system 9 for learning predictive model 110
3.1 Outline Next, a method for generating the prediction model 110 included in the integrated health prediction program 100 by machine learning will be described with reference to FIG. Here, it is assumed that the model learning program 900 generates the prediction model 110 . Note that the model learning program 900 may be implemented as a single program, or may be implemented as a plurality of programs that can operate in cooperation by inputting and outputting data.

Note that the integrated health prediction program 100 illustrated in FIG. 1 includes six types of prediction models 110, but these prediction models differ only in the health data 954 of the teacher data 950 to be used. can be generated to Below, the case of generating the menopausal prediction model 110a relating to women's health will be mainly described.

3.2 Teacher data 950
The model learning program 900 performs machine learning using teacher data 950 including various data collected from a large number of subjects, and constructs the prediction model 110 . The teacher data 950 includes purchase data 952 regarding the purchasing behavior of each subject and health data 954 which is data regarding the health of those subjects.

Purchase data 952 relates to information similar to that illustrated with reference to FIG. That is, in this embodiment, for each ID of the subject, items 150a related to categories in the JICFS classification, and performance data 150b, which is numerical data such as the purchase amount, purchase quantity, and number of purchases during the period, are stored in the purchase data 952. can contain. Since food and/or beverages are suitable as commodities associated with the health conditions of subjects and customers, the purchase data 952 preferably includes the purchase history of food and/or beverages.

Health data 954 describes the health status of the subject. The health condition may be a condition related to subjective symptoms or a condition recognized as a result of measurement. That is, at least one of health awareness, mental health, cognitive function, and physical examination results can be used as the health data 954 . In the health data 954 in this embodiment, the judgment value for each condition segment of the subject is described as the health condition of the subject. The condition segment is defined for each combination of health condition type and level (in other words, magnitude). The type and level of health condition referred to in the present embodiment are the type and level of mental and physical disorders that the corresponding individual has. The judgment value can represent an index indicating the state of health in 2 classes, 3 classes, or the like. It should be noted that the "type and level of health condition" referred to here does not involve medical rigor, but can also be understood as the type and level of health concerns that an individual has.

Now, as an example, consider the case where the model learning program 900 generates the menopausal prediction model 110a and uses the symptoms of the Brief Menopausal Index SMI to determine the level of menopause. In this case, the health data 954 may include a label indicating the symptoms of the Simplified Menopausal Index SMI to which the subject belongs. For example, the Simplified Menopause Index SMI can be expressed as two classes. Specifically, it is "H" indicating that menopausal symptoms are relatively large, and "other than H". "Other than H" can be subdivided into "M", which indicates relatively moderate symptoms, and "L," which indicates relatively minor symptoms (in other words, symptoms can be expressed as a total of three classes). is. Below, the subject layer corresponding to "H" is called "H layer".

It should be noted that the labeling indicating the degree of symptoms in the health data 954 may be further subdivided into four or more classes instead of two or three classes, or alternatively, instead of these classes, there is a possibility of corresponding to menopausal symptoms. It is also conceivable to use a numerical value that indicates the nature.

That is, the training data 950 includes information on each product category of each subject, purchase data 952 including information on the purchase amount, purchase quantity, number of purchases, etc. during a predetermined period, and label information indicating health conditions at least. It is associated with the health data 954 included. Here, the model learning program 900 includes, for example, women's health condition (menopausal symptoms), sleep condition (disturbed sleep), lifestyle-related disease (lack of physical activity), water condition (lack of water), immune condition (immune degree) and nutritional status (degree of malnutrition), information corresponding to each of these prediction models 110 should be included in the health data 954 . In machine learning in supervised learning, teacher data 950 is broadly classified into explanatory variables and objective variables. In this embodiment, the purchase data 952 correspond to explanatory variables, and the health data 954 correspond to objective variables. In this embodiment, only the purchase data 952 is used as explanatory variables, but the explanatory variables are not limited to this. For example, it is conceivable to add gender and age, which are subject attributes, to explanatory variables.

Hereinafter, if appropriate health data 954 is prepared, the model learning program 900 can be used for sleep conditions, lifestyle-related diseases, immune conditions, water conditions, nutritional conditions, dementia, oxygen utilization, blood vessel health, hair health, intestinal health. Environment, promotion of exercise effects, appetite enhancement, eye health, skin health, oral health, mental health, depression, stress, headache, stiff neck, ear health, joint health, body temperature homeostasis, fatigue, bone health , male menopause, hypertension, weather pain, etc., and the like. Below, in particular, women's health conditions (menopausal symptoms), sleep conditions (disturbed sleep), lifestyle-related diseases (lack of physical activity), hydration conditions (hydration levels), immune conditions (immune depression), and nutritional conditions Examples of indices that can be used as indicators of (degree of undernutrition) are shown below.

3.2.1 Health status of women (menopausal symptoms)
As an index indicating the degree of climacteric symptoms, for example, as described above, the simplified climacteric index SMI can be used. The Simplified Menopausal Index SMI is obtained by asking each subject to fill out relevant items corresponding to the degree of various listed symptoms in advance and scoring the results. The simplified climacteric index SMI is considered to be an index reflecting symptoms peculiar to Japanese menopausal women. Therefore, depending on the country or region, other indices related to menopausal symptoms, such as the Menopause Rating Scale, The Kupperman index, the Green Climatic Scale, PSST （Ｔｈｅ ｐｒｅｍｅｎｓｔｒｕａｌ ｓｙｍｐｔｏｍｓｓｃｒｅｅｎｉｎｇ ｔｏｏｌ）、ＭＲＳ（Ｍｅｎｏｐａｕｓｅ ｒａｔｉｎｇ Ｓｃａｌｅ）、ＷＨＱ（Ｔｈｅ Ｗｏｍｅｎ'ｓ Ｈｅａｌｔｈ Ｑｕｅｓｔｉｏｎｎａｉｒｅ）、ＶＡＳ（Ｖｉｓｕａｌ ａｎａｌｏｇｕｅ ｓｃａｌｅ）、ＨＦＲＤＩ（Ｈｏｔ Ｆｌａｓｈ Ｒｅｌａｔｅｄ Ｄａｉｌｙ Ｉｎｔｅｒｆｅｒｅｎｃｅ Ｓｃａｌｅ）、ＨＦＣＳ（Ｈｏｔ Ｆｌａｓｈ Ｃｏｍｐｏｓｉｔｅ Ｓｃｏｒｅ）、及びAppropriate use of one or more of MENQOL (Menopause-Specific Quality of life) and Japanese women's menopausal symptom evaluation chart is also conceivable.

In addition to menopausal symptoms, premenstrual syndrome (PMS, PMDD) symptoms are also considered as women's health conditions. That is, in order to predict premenstrual syndrome, the health prediction model 110 may be constructed by machine learning using an index indicating the degree of symptoms of premenstrual syndrome as an objective variable and the purchase data 952 as explanatory variables.

Also, the health data 954 may include numerical data of these indices, or may include only labels indicating the degree of symptoms from the numerical values of these indices. This also applies to the indices of sleep state, lifestyle-related diseases, and the like, which will be described later.

3.2.2 Sleep state (disturbed sleep)
The index indicating sleep may be an index indicating deterioration of sleep state and/or sleep rhythm, or inappropriate sleep duration. It may be a subjective symptom of each control acquired by a questionnaire or the like, or it may be a measured value recognized as a result of measuring each control. Sleep status includes Athens Sleep Assessment Scale (AIS), Pittsburgh Sleep Quality Index (PSQI), 3 Dimensional Sleep Scale (3DSS), Insomnia Severity Index (Insomnia Severity Index). : ISI) can be used. The degree of symptoms is analyzed individually and in combination for each item, and the health condition of the consumer is assigned and classified into categories. The sleep rhythm is grasped by using the Munich ChronoType Questionnaire (μMCTQ), obtaining the median sleep time on weekdays and holidays, and calculating the social jet lag (SJL) from the difference. be able to.

3.2.3 Lifestyle-related diseases (lack of physical activity)
The exercise status index may include one or more selected from lack of exercise, obesity, and changes in body weight over a certain period of time. More specifically, one or more of each subject's body weight, BMI (Body Mass Index), number of steps, and/or metabolic syndrome diagnostic reference value may be used. Metabolic syndrome diagnostic criteria include waist circumference, thresholds established for diagnosis of hyperglyceridemia and/or low HDL cholesterol, maximum and/or minimum blood pressure, and fasting hyperglycemia. Any one or more of thresholds established for diagnosis.

In addition to the purchase data 952, in addition to the purchase data 952, machine learning is performed by including one or more information such as the subject's smoking status and sedentary behavior as explanatory variables when performing machine learning to generate a prediction model for lifestyle-related diseases and exercise conditions. you can go

3.2.4 Moisture status (moisture deficit)
Indices of hydration status include serum Na value, BUN/creatinine ratio, and/or urine osmotic pressure, urine specific gravity, urine color, dehydration evaluation scale, and subjective symptoms associated with dehydration. These information can be obtained in a manner similar to the menopausal symptoms example above. For example, physical changes experienced by each of a plurality of controls for a certain period of time, for example, one year, are obtained from clinical examination values obtained from physical examinations and complete medical examinations, or from health questionnaires separately conducted. On the other hand, the subjective symptoms associated with dehydration can be acquired through a questionnaire or the like. The degree of symptoms can be grasped from at least one or more of serum Na value, urine osmotic pressure, urine specific gravity, BUN/creatinine ratio, water intake, amount of physical activity, and urine color.

Based on these indicators, it is possible to find epidemiological information in the literature for conditions that are experiencing or may experience worsening health conditions. For example, it has been found that consumers with serum Na levels above 142 mEq/L have a higher risk of developing cognitive impairment and hypertension than those with serum Na levels below 142 mEq/L, and are assessed to be at high health risk. It is possible to

In addition to the purchase data 952, in addition to the purchase data 952, the amount of water and alcohol consumed, the amount of physical activity, the urine color, the urine osmotic pressure, the specific gravity of the urine, the serum Na value, the serum Na value, and the BUN /creatinine ratio, machine learning may be performed by including one or more information as an explanatory variable.

3.2.5 Immune status (degree of immunosuppression)
For example, susceptibility to and/or frequency of colds and cold-like symptoms can be used as indicators of immune status. The susceptibility and frequency of catching a cold may be the subjective symptoms of each control acquired by a questionnaire or the like, or may be acquired from the history of consultation at a medical institution. The degree of symptoms is aggregated into three or more categories, and the degree of symptoms is susceptibility to colds, SIgA concentration, allergic symptoms, oral environment/immune status, exercise status, stress status, and severe sleep symptoms. , moderate sleep symptoms, mild sleep symptoms, cold-like symptoms.

3.2.6 Nutritional status (degree of malnutrition)
The index representing the nutritional status may be the subjective symptoms of each control acquired by a questionnaire or the like, or may be a measured value recognized as a result of measuring each control. The degree of symptoms can be at least one of the presence or absence of subjective symptoms associated with malnutrition, diversity score of food intake (DVS), simplified nutritional appetite question (SNAQ), and BMI (Body Mass Index) .

3.3 Machine Learning Method The model learning program 900 according to this embodiment uses logistic regression as a machine learning method. However, logistic regression is an example and other techniques such as random forests, decision trees, gradient boosting, support vector regression, linear regression, Partial Least Squares (PLS) regression, Gaussian process regression, Adopting a neural network or the like is also conceivable. Also, when logistic regression is used, the accuracy may be further improved by using the L1 normalization method, or logistic regression (ridge regression) using the L2 normalization method is adopted instead of the L1 normalization method. You may Elastic nets that employ both L1 and L2 regularization methods can also be used.

Logistic regression is a method of determining whether or not an input corresponds to a certain event by calculating the probability that the input corresponds to a certain event (or the probability that the input does not correspond to a certain event). A logistic regression model is described as an example of the following equation (1) called a logit function. In Equation (1), the left side is the natural logarithm. Also, i is the i-th subject, _{p is the probability that the event of the objective variable occurs, b 1 , b 2 . xn} indicates an explanatory variable, respectively.

Assuming that the right side of Equation (1) is z, Equation (1) can be transformed into Equation (2) representing probability p.

式（２）はシグモイド関数と呼ばれ、ロジスティック回帰モデルにおける活性化関数である。ｚの定義における偏回帰係数ｂ_１、ｂ_２…ｂ_ｎは重みと呼ばれ、ｂ_０はバイアス項である。ｚをシグモイド関数に入力すると、０～１の値、すなわち確率値が算出される。

Equation (2) is called the sigmoid function and is the activation function in the logistic regression model. The partial regression coefficients b ₁ , b _{2 .} . . b _n in the definition of z are called weights and b ₀ is the bias term. Inputting z into the sigmoid function yields a value between 0 and 1, ie a probability value.

Here, the sigmoid function obtained by the equation (2) is set as "φ(z)", and the output value is classified into an output y indicating one of two values. Equation (3) means that if φ(z) is 0.5 or more, it is classified into class 1, and if it is less than 0.5, it is classified into class 0. Note that 0.5 is an example, and a value other than 0.5 may be used as the threshold. In other words, formula (3) means that if z in formula (2) is 0 or more, it is classified into class 1, and if z in formula (2) is less than 0, it is classified into class 0. .

Next, we introduce the “likelihood” used for learning in logistic regression. Likelihood means the likelihood of a condition in terms of results. A likelihood function L representing likelihood is represented by the following equation (4). Note that P represents a probability value.

The likelihood function L indicates the probability of correctly determining all events. By obtaining the weight that maximizes the likelihood, the probability of the event to be predicted can be output more accurately. Specifically, the likelihood function L is multiplied by (-1) to reverse the government of the likelihood function. This positive/negative-reversed likelihood function is the error function in logistic regression. In order to obtain the weights that minimize the error function, that is, the partial regression coefficients b ₁ , b ₂ . . _{. b n ,} the error functions b ₁ , b 2 . Thus, learning of logistic regression is performed.

As the likelihood function, a logarithmic likelihood function using the natural logarithm of Equation (4) may be used. The optimum weight can be found by minimizing the function obtained by multiplying the logarithmic likelihood function by (-1) and reversing the sign.

Computational algorithms for the logistic regression described above are well known, and software applications for performing machine learning with such algorithms are readily available. Although the principle is as described above, if available software is used, even a person who does not know the details of a specific analysis method using formulas can understand the machine using logistic regression according to this embodiment. Learning is possible.

3.4 Method of Determining Product Categories Used for Health Prediction corresponds to the explanatory variable x in .

As described above, the health data 954 included in the teacher data 950 in the present embodiment has "H" indicating whether each symptom such as menopausal symptom or sleep disturbance is classified into the H layer or assigned to the L layer. or "L" (or "H", "M", "L") labels are assigned to each subject. In the above equation (3), the likelihood function can be obtained by classifying the layer so that "1" is output for the H layer and "0" is output for the layer other than the H layer. Elements that make up the explanatory variable x are not limited to the purchase data 952 described above. For example, explanatory variables may include data indicating age groups and data indicating gender.

By performing machine learning using teacher data 950 on a large number of subjects and constructing a prediction model, it is possible to determine product groups (categories) that H-tier customers often purchase.

FIG. 10 shows the product categories selected (often purchased) by many test subjects in the H class and other classes regarding menopausal symptoms. The upper part of FIG. 10 shows the product categories selected by most of the H-tier subjects, and the lower part shows the product categories selected by subjects other than the H-tier subjects.

In addition, FIG. 11 shows the relationship between menopausal symptoms found in items purchased by H-tier subjects regarding menopausal symptoms and health issues. The numerical values in the rightmost column of FIG. 11 indicate the extent (deviation value) indicating the ease of selection by women with menopausal symptoms. This number is quoted from "Deviation value by symptom for items that you want to deal with even if it costs money - Research project report on women's health and beauty lifestyle understanding - Behavior observation survey results".

From these results, it can be seen that the H-level subjects regarding menopausal symptoms actively select (purchase) specific products. The H-tier subjects purchased products for dry skin and body, and for indefinite complaints.

FIG. 12 shows an example of a product group (product category) for women who belong to the H tier and have menopausal symptoms. In this way, it is possible to determine the product categories that H-tier subjects frequently purchase compared to non-H-tier subjects. It can be estimated that the subject is likely to belong to the H stratum in menopausal symptoms. In the same way, it is possible to determine the product category that subjects who do not belong to the H stratum frequently purchase compared to the H stratum subjects. It can be estimated that there is a high possibility that it does not belong to the H layer.

Therefore, the model learning program 900 in this embodiment builds the predictive model 110 using the product category data that can contribute to the estimation that the purchase data 952 belongs to the H tier or does not belong to the H tier. By doing so, it is possible to estimate the customers belonging to the H layer with high accuracy.

3.5 Configuration of Model Learning Program 900 The functional configuration of the model learning program 900 will be described below with reference to FIG. FIG. 9 is a diagram showing the functional configuration of the model learning program 900. As shown in FIG. The model learning program 900 has an input section 910 , a product category determination section 912 , a learning section 914 and an output section 916 . Note that the model learning program 900 may be implemented as a single program, or may be implemented as a plurality of cooperating programs.

Input unit 910 receives input of teacher data 950 . The input unit 910 may read the teacher data 950 from a storage medium such as an HDD or SSD built into the information processing device (computer) on which the model learning program 900 is executed, or may read the teacher data 950 via a network such as a LAN or the Internet. Input of the teacher data 950 may be received from an external information processing device or the like to be connected.

The product category determination unit 912 selects the product category of the purchase data 952 used for machine learning of the prediction model 110 to be learned. Since the product category selection method was described in 3.4 above, detailed description is omitted here. Note that when generating a plurality of predictive models 110 corresponding to a plurality of symptoms, the product category determining unit 912 can select different product categories for each predictive model 110 .

As described above, it is conceivable to create the prediction model 110 using the entire purchase data 952 without narrowing down the product category during the machine learning of the prediction model 110 . In that case, the product category determination unit 912 is unnecessary.

The learning unit 914 performs machine learning using the purchase data 952 of the product category determined by the product category determination unit 912 to generate the prediction model 110 . Since the method of generating the prediction model 110 has been described in 3.3 above, detailed description thereof will be omitted here.

The output unit 916 outputs the prediction model 110 generated by the learning unit 914 to another information processing device (for example, the information processing device 800 described with reference to FIG. 8) connected via an internal storage medium or a network. etc. As an output method, it is conceivable to output the function itself constituting the prediction model 110, or to output it as a parameter of the function constituting the prediction model 110. FIG.

3.6 Flow of Processing The flow of processing of the model learning program 900 will be described below with reference to FIG. FIG. 13 is a flow chart showing the procedure for creating the prediction model 110, which is executed by the model learning program 900. As shown in FIG.

The input unit 910 of the model learning program 900 acquires teacher data 950 on many subjects used for learning (S1301). The teacher data 950 includes purchase data 952 indicating the subject's purchase history and health data 954 indicating the health condition (degree of symptoms). As described above, the purchase data 952 can include, for example, the JICFS classification (classification code) and product category, and numerical data such as the number of purchases, purchase amount, and purchase quantity corresponding thereto.

Next, the product category determination unit 912 determines the JICFS classification and product category used to generate the prediction model 110 (S1303). The determination method is as described in 3.4 above. Note that when the prediction model 110 is created using the entire purchase data 952 without narrowing down the product category, the process of S1303 is unnecessary.

The learning unit 914 performs machine learning using the purchase data 952 as an explanatory variable and the health data 954 as an objective function (S1305). Since the machine learning method has been described with specific examples in 3.3 above, the description is omitted here.

The output unit 916 outputs the prediction model 110 predicted by the learning unit 914 in S1305 to an internal storage medium or other information processing apparatus connected via a network. As described above, the predictive model 110 may be output as the function itself constituting the predictive model 110 or in the form of a parameter or the like.

3.7 Hardware Configuration A specific example of the hardware configuration of the information processing apparatus 1400 in which the model learning program 900 is executed will be described with reference to FIG. FIG. 14 is a hardware configuration diagram of an information processing device 1400 on which the model learning program 900 is executed. The information processing device 1400 uses the model learning program 900 to generate the prediction model 110 that predicts the customer's menopausal symptoms, sleep state, and the like.

The information processing device 1400 can be implemented as a generally available computer system, and may be installed at the business office of the business operator that operates the health prediction system 1, or may be provided as a cloud service, etc., as described above. It is similar to the information processing device 800 . In this embodiment, the information processing device 800 that performs health prediction and the information processing device 1400 that generates the prediction model 110 are mainly described as separate devices, but both are implemented as the same device. good too.

Information processing apparatus 1400 includes control unit 1410 , communication I/F unit 1420 , and storage device 1430 .

Control unit 1410 may include a CPU, ROM, RAM, and the like. Control unit 1410 can execute model learning program 900 stored in storage device 1430 . Thereby, the information processing apparatus 1400 can execute various processes related to the construction of the prediction model 110 described above, in addition to the function as a general computer. Note that the arithmetic circuit included in the control unit 1410 may not be a CPU, and may be realized by various processors such as MPU and GPU. Moreover, it may be realized by a plurality of processors instead of one processor.

The communication I/F unit 1420 and the input I/F unit 820 communicate according to standards such as an Ethernet (registered trademark) communication terminal, a USB (registered trademark) terminal, IEEE 802.11, 4G, or 5G. It can be realized by a communication circuit that performs Communication I/F section 1420 can be connected to a communication network such as an intranet or the Internet, and receives teacher data 950 prepared by a business operator who operates learning system 9 . In addition, information processing apparatus 1400 may directly communicate with another device via communication I/F section 1420, or may communicate via an access point or the like.

The storage device 1430 is a storage medium that stores computer programs and data necessary for operating the information processing device 1400 . Storage device 1430 may be, for example, an HDD or SSD. The storage device 1430 may include a temporary storage element configured by a RAM such as DRAM or SRAM, and may function as a work area for the control section 1410 . The storage device 1430 stores the model learning program 900 and the prediction model 110 generated by the model learning program 900 (including tables and/or parameter groups defining the prediction model 110).

4 Effect of this embodiment In the health prediction system 1 according to this embodiment, the integrated health prediction program 100 can predict a plurality of health risks of the customer from the purchase data 150 of the customer. Then, based on individual prediction results, it is possible to make detailed proposals for improving and/or maintaining current or future health conditions for each customer.

5 Supplementary Note It should be noted that the configurations of the above-described embodiments may be combined or partly replaced. Also, the configuration of the present invention is not limited to the above-described embodiments, and various modifications may be made without departing from the gist of the present invention.

For example, in the above-described embodiment, the integrated health prediction program 100 includes six types of health prediction models, but including six types is merely an example. Two or more types, less than six types, or seven or more types of health predictions may be included. Generally described, the integrated health prediction program 100 may include predictive models 110 for each health condition type 1 to type N (where N is an integer greater than or equal to 2). Then, the health prediction model 110 for the k-th type k-th (1≤k≤N) health symptom is input with the purchase data 150 regarding the product category to which at least one product belongs, and the degree of the k-th type health symptom is calculated. It is sufficient if it is constructed as a prediction model to be output. The health prediction model 110 regarding the k-th type (1≤k≤N) health condition uses purchase data 952 related to the purchase history of each of a plurality of subjects as an explanatory variable, and uses the health condition collected from each of the subjects as an explanatory variable. It can be constructed by machine learning using the health data 954 representing the degree of as an objective variable.

In addition, the numerical values of the upper and lower limits of the numerical range defining the class when the degree of health symptoms is represented by numerical values are examples, and can be changed as appropriate. The number of defined classes can also be arbitrarily determined by those skilled in the art.

6. Other Embodiments It is also conceivable to appropriately combine part or all of the embodiments described above with the embodiments described below. Details will be described below.

6.1. Overview In the technology disclosed in the present application, briefly, at least one information processing device acquires target behavior data that identifies a history of behavior performed by a user to be inferred (target user). By inputting the target behavior data into the estimation model, data for identifying the target user's health condition (health condition data) can be output from the estimation model.

Also, at least one information processing device may determine at least one target product and/or at least one target service corresponding to the target user's health condition data from among a plurality of products and/or a plurality of services. can. Further, the at least one information processing device may output the at least one target service thus determined and/or information identifying the at least one target service.

6.2 Configuration of Communication System Such a technique can be implemented using a communication system as illustrated in FIG. FIG. 15 is a block diagram showing an example of the configuration of a communication system according to one embodiment.

As shown in FIG. 15, a communication system 1500 according to an embodiment includes, for example, at least one server device (information processing device) 10 connectable to the communication network 2 and at least one terminal connectable to the communication network 2. A device (information processing device) 20 can be included. Although two server devices 1510A and 1510B are illustrated as at least one server device 1510 in FIG. 15, any number of server devices 1510 can be used. Similarly, although two terminal devices 1520A and 20B are illustrated as at least one terminal device 1520 in FIG. 15, any number of server devices 1510 can be used.

Each server device 1510 can be connected to at least one other server device 1510 and/or at least one terminal device 1520 via the communication network 2 . Also, each terminal device 1520 can be connected to at least one other terminal device 1520 and/or at least one server device 1510 via the communication network 2 .

Each server device 1510 can be any information processing device. Such information processing devices can include, for example, but are not limited to, personal computers, workstations, supercomputers, mainframes, and the like. Each terminal device 1520 may be any information processing device. Such information processing devices can include, but are not limited to, mobile phones, smart phones, tablets, personal computers, workstations, personal digital assistants, and the like.

In one example, at least one server device 1510 provides a service for estimating the health status of any user ("estimation service") and/or at least one target product corresponding to the health status of any user. It may be an information processing device managed by an operating company that operates a proposed service (“proposed service”). Hereinafter, for convenience, the term "estimation service, etc." may be used as a term indicating the estimation service and/or the proposal service.

In one example, at least one server device 1510 and/or at least one terminal device 1520 may be an information processing device managed by a company user and/or an individual user who receives estimation services and the like from the operating company.

The operation of acquiring target behavior data described above can be performed by at least one server device 1510 and/or at least one terminal device 1520 .

The estimation model described above can be generated by at least one server device 1510 and/or at least one terminal device 1520 . The estimation model thus generated can be held by at least one server device 1510 and/or at least one terminal device 1520 . The estimation model held in this way can be used by at least one server device 1510 and/or at least one terminal device 1520 .

At least one server device 1510 and/or at least one terminal device 1520 may perform the operation of causing the health condition data to be output from the estimation model described above.

The operation of determining at least one target product and/or at least one target service described above may be performed by at least one server device 1510 and/or at least one terminal device 1520 . The operations of outputting at least one target service and/or data identifying at least one target service described above may be performed by at least one server device 1510 and/or at least one terminal device 1520 .

The communication network 2 includes a mobile phone network, a wireless network, a fixed phone network, the Internet, an intranet, a local area network (LAN), a wide area network (WAN), and/or an Ethernet (registered trademark) network. can be included without limitation. The wireless network is limited to RF connections via, for example, Bluetooth®, WiFi (such as IEEE 802.11a/b/n), WiMax, cellular, satellite, laser, infrared, etc. can be included without

6.3 Hardware Configuration of Each Information Processing Device Next, an example of a hardware configuration of each information processing device (server device 1510, terminal device 1520) will be described.

(1) Hardware Configuration of Server Device 1510 FIG. 16 is a block diagram showing an example of the hardware configuration of the server device 1510 shown in FIG. (described with reference to terminal 1520 to do so).

As shown in FIG. 16, the server device 1510 includes a central processing unit 11, a main storage device 12, an input/output interface device 13, an input device 14, an auxiliary storage device 15, and an output device 16. can be done. These devices are connected to each other by a data bus and/or a control bus.

The central processing unit 11 is called a “CPU” and can perform operations on instructions and data stored in the main memory 12 and store the results of the operations in the main memory 12 . Furthermore, the central processing unit 11 can control the input device 14, the auxiliary storage device 15, the output device 16, etc. through the input/output interface device 13. FIG. Server device 1510 may include one or more such central processing units 11 .

The main storage device 12 is referred to as "memory", and receives commands and data from the input device 14, the auxiliary storage device 15, the communication network 2, etc. (terminal device 1520, etc.) via the input/output interface device 13, and The calculation results of the central processing unit 11 can be stored. The main memory 12 includes volatile memory (e.g., registers, cache, random access memory (RAM)), nonvolatile memory (e.g., read-only memory (ROM), EEPROM, flash memory), and storage (e.g., hard disk computer readable media such as drives (HDD), solid state drives (SSD), magnetic tapes, optical media), etc., without limitation. As will be readily understood, the term "computer-readable recording medium" refers to data storage media such as memory and storage, rather than transmission media such as modulated data signals or transient signals. It can contain media.

The auxiliary storage device 15 is a storage device having a larger capacity than the main storage device 12 . The auxiliary storage device 15 stores instructions and data (computer programs) that constitute a specific application or the like, and is controlled by the central processing unit 11 to input/output these instructions and data (computer programs). It can be transmitted to the main storage device 12 via the interface device 13 . The auxiliary storage device 15 can include, but is not limited to, a magnetic disk device and/or an optical disk device.

The specific applications referred to herein include operating systems, various software applications (including web browsers), dedicated applications run to provide the Estimation Service, etc., dedicated applications run to receive the Estimation Service, etc. applications such as, but not limited to,

The input device 14 is a device that takes in data from the outside, and can include, without limitation, a keyboard, touch panel, button, mouse and/or sensor (microphone, camera) and the like.

Output devices 16 may include, without limitation, display devices, touch panels, and/or printer devices.

In such a hardware configuration, the central processing unit 11 sequentially loads the instructions and data (computer programs) constituting the specific application stored in the auxiliary storage device 15 into the main storage device 12, and loads them. It is possible to operate on the instructions and data that have been processed. Thereby, the central processing unit 11 controls the output device 16 via the input/output interface device 13, or controls another device (for example, another server device 1510) via the input/output interface device 13 and the communication network 2. and/or the terminal device 1520 or the like) can transmit and receive various information (data).

In this way, the server device 1510 performs the operations related to providing the estimation service and/or receiving the estimation service and the like (FIGS. 17 and 17) by executing the specific installed application. 26, etc.) can be performed. Additionally or alternatively, the server device 1510 may execute a browser to access another server device 1510 to provide and/or receive an estimation service or the like. Related operations and the like (various operations and the like to be described later with reference to FIGS. 17 and 26, etc.) can also be performed.

It should be noted that server device 1510 may include one or more microprocessors and/or graphics processing units (GPUs) in place of or in conjunction with central processing unit 11 .

(2) Hardware Configuration of Terminal Device 1520 As shown in parentheses in FIG. As shown in FIG. 16, the terminal device 1520 includes a central processing unit 21, a main memory device 22, an input/output interface device 23, an input device 24, an auxiliary memory device 25, and an output device 26. can be done. These devices are connected to each other by a data bus and/or a control bus. Each of these devices is as described with respect to server device 1510 .

The specific applications stored in the secondary storage device 25 and executed by the central processing unit 21 include the operating system, various software applications (including web browsers), dedicated applications executed to provide estimation services, etc. etc., without being limited thereto. Specific applications may also include specialized applications that are run to provide estimation services and the like.

In such a hardware configuration, the central processing unit 21 sequentially loads the instructions and data (computer programs) constituting the specific application stored in the auxiliary storage device 25 into the main storage device 22, It is possible to operate on the instructions and data that have been processed. As a result, the central processing unit 21 controls the output device 26 via the input/output interface device 23, or via the input/output interface device 23 and the communication network 2, other devices (for example, the server device 1510 and and/or other terminal devices 1520, etc.) can transmit and receive various information (data).

In this way, the terminal device 1520 receives provision of the estimation service or the like and/or performs operations related to providing the estimation service or the like (Figs. 26, etc.) can be performed. Additionally or alternatively, the server device 1510 may execute a browser to access another server device 1510 to provide and/or receive an estimation service or the like. Related operations and the like (various operations and the like to be described later with reference to FIGS. 17 and 26, etc.) can also be performed.

Note that terminal device 1520 may include one or more microprocessors and/or graphics processing units (GPUs) in place of or in addition to central processing unit 21 .

6.4 Operations Performed by Communication System 1500 to Provide Sleep State Data Regarding Sleep States A specific example will be further described with reference to FIG. FIG. 17 is a flow diagram illustrating an example of operations performed by the communication system 1500 shown in FIG.

(1) Step 1000
First, in step (hereinafter referred to as “ST”) 1000, an information processing device, for example, a server device 1510A managed by a company that operates an estimation service or the like, generates a plurality of sets of estimation models. Teacher data can be acquired and stored.

Each set of teacher data constituting a plurality of sets of teacher data includes data identifying the history of actions performed by one sample user corresponding to this set ("sample action data") and and data identifying sleep states of one sample user ("sample sleep state data"). For example, the first set of teacher data includes sample action data identifying the history of actions performed by one sample user (Mr. A) corresponding to the first set, and sample action data identifying the sleeping state of Mr. A. sleep state data, wherein the second set of teacher data includes sample behavior data identifying a history of behavior performed by one sample user (Mr. B) corresponding to the second set; and sample sleep state data identifying Mr.'s sleep states.

(1A) Sample behavior data Each sample behavior data may be data that identifies at least one (one or more) behavior among a plurality of predetermined behaviors. Here, each of the plurality of predetermined actions may be actions that can at least partially have a positive or negative effect on the sleep state of the person who performed the action. Such predetermined behaviors can include, for example, but not limited to, those exemplified below.
(Examples of behaviors that can at least partially positively affect a person's state of sleep)
・Performed aerobic exercise for 1 hour or more at a pace of 2 days or more per week.
・I got up at a certain time every day.
-Limited to 2 cups of caffeine-containing beverages per day.
・Working hours per day were less than 8 hours.
- Purchased 10 or more bottles of coffee (this behavior can be identified by data including, for example, the product category to which the coffee belongs and its purchase quantity).
(Examples of behaviors that can at least partially adversely affect a person's state of sleep)
- Did aerobic exercise for 1 hour or more at a pace of 1 day or less per week.
-Woke up at different times every day.
・Drinking 3 or more cups of caffeine-containing beverages per day.
・Working hours per day were 13 hours or more.
• Purchased 5 or more bottles of milk (this behavior can be identified, for example, by data containing the product category to which the milk belongs and its purchase quantity).

Each sample behavior data may be data identifying each of at least one behavior among a plurality of predetermined behaviors exemplified in this way.

Also, in one example, since the product purchased by the user and/or the service used by the user may tend to have a positive or negative effect on the sleep state of the user, the plurality of predetermined actions are , the product purchased by the sample user (or the category to which the product belongs), and the service used by the sample user (or the category to which the service belongs). This type of action can be generated using, for example, POS (Point Of Sales) data held by retail stores (drug stores, supermarkets, convenience stores, etc.).

In one example, each of a plurality of predetermined behaviors is assigned unique identification data (such as letters, numbers, or combinations thereof), and each sample behavior data may include such identification data. can. For example, when using POS data, each of the plurality of sample behavior data includes the JICFS classification (classification code), product category, purchase amount, purchase quantity, number of purchases, gender, and age corresponding to the purchased product. , and/or the date of purchase, etc., without limitation.

For example, a server device 1510A managed by a company that operates an estimation service or the like, which is an information processing device, can acquire such sample behavior data by, for example, at least one of the methods exemplified below. can.
- The information processing device, via the communication network 2, at least one other server device 1510 (for example, another server device 1510 that stores POS data installed in a drugstore or the like, or can access this POS data) POS data is received from another server device 1510).
- The information processing device sends an e-mail or a web page that describes the answers to a predetermined questionnaire (questionnaire about actions taken by the sample users) via the communication network 2 to at least one terminal device 1520 and/or at least one Received from server device 1510 .
- The information processing device receives, via the communication network 2, the content of the responses from at least one server device 1510 that has collected responses from a plurality of users to the predetermined questionnaire.
- The information processing device receives the contents of the answers from a plurality of users to the predetermined questionnaire via a recording medium (USB memory, DVD-ROM, etc.).

(1B) Sample sleep state data The sample sleep state data included in each set of teacher data can include, without limitation, at least one of the data exemplified below.
data identifying the sleep disposition of one sample user corresponding to the set ("sample disposition data");
- Data identifying the sleep rhythms of one sample user corresponding to the set ("sample rhythm data")
- Data identifying the sleep duration of one sample user corresponding to the set ("Sample Time Data")
- Data identifying the sleep category to which (the sleep of) one sample user corresponding to the set belongs ("sample sleep category data")

First, the sample disposition data can be data that identifies whether the sample user's sleep quality is good (eg, how good) or poor (eg, how bad). In one example, the sample disposition data may be data identifying an overall score for the sleep quality of the sample user. In this case, the sample user's sleep quality may be data identifying a score calculated by any of the methods exemplified below, for example.

-A score calculated by the method described in the Athens Insomnia Scale (AIS) (see FIG. 18) based on the sample user's answers to the questionnaires described in the Athens Insomnia Scale, or any method for this score Corrected score obtained by calculation (addition, subtraction, multiplication, division, etc.) Pittsburgh sleep question based on the sample user's answers to the questions described in the Pittsburgh sleep questionnaire (PSQI) (see FIGS. 19A to 19C) A score calculated by the method described in the vote, or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, etc.) Three-dimensional sleep scale (3DSS) (FIGS. 20A and 20B Reference) based on the contents of the sample user's answers to the questionnaire items described in), the score calculated by the method described in the three-dimensional sleep scale, or this score is calculated by any method (addition, subtraction, multiplication, division, etc.) Corrected score obtained by Insomnia Severity Questionnaire (ISI) (see Figure 21) based on the sample user's answers to the questions described in the Insomnia Severity Questionnaire by the method described in Calculated score or modified score obtained by calculating this score by any method (addition, subtraction, multiplication, division, etc.) based on the score calculated by the method described in the questionnaire or scale, or a modified score obtained by calculating this score by any method

Second, sample rhythm data can be data that identifies whether the sample user's rhythm quality is good (eg, how good) or bad (eg, how bad). In one example, the sample rhythm data may be data identifying a score evaluating the sleep rhythm of the sample user.

In this case, the sleep rhythm of the sample user is, for example, a social jet lag (or social jet Lag calculated in any manner). Here, for example, if a certain sample user sleeps during weekdays from midnight to 6:00 a.m., the central time is 3:00 a.m. If it is 3:00 to 11:00 am, then the median time is 7:00 am, so the social jet lag is 4 hours. In general, it can be said that the smaller (or greater) the social jet lag, the better (or worse) the sample user's sleep rhythm.

The sample user's sleep hours on weekdays and holidays, in one example, can be extracted from the sample user's answers to the questionnaire described in the Munich Chronotype Questionnaire (MCTQ) (see FIGS. 22A and 22B). However, in another example, it may be obtained from the sample user via e-mail, web page, or the like.

Third, the sample time data can be data that identifies the sleep hours (sleep hours per day) of the sample users. A sample user's sleep time can be extracted, in one example, from the sample user's responses to questionnaires listed in the Munich Chronotype Questionnaire (MCTQ) (see FIGS. 22A and 22B), and in another example, It can be obtained from the sample user via e-mail, web page, or the like.

Fourth, the sample sleep category data can be data identifying the sleep category to which the sample user's (sleeps) belong. As an example, the sample sleep category data may be data identifying the category to which the sample user belongs, determined based on the sample user's sample disposition data and the sample user's sample rhythm data. . In this case, the sleep category can be defined by, for example, the following method.

FIG. 23 is a diagram showing an example of sleep category data used in communication system 1500 shown in FIG. As shown in FIG. 23, on one of the vertical and horizontal axes (here, the horizontal axis), the score identified by the sample property data is arranged, and on the other of the vertical and horizontal axes (here, the vertical axis), the sample A value identified by the rhythm data may be placed. On the one axis, at least one (here, one) threshold may be set for the score identified by the sample property data, namely the threshold A1 (6 points). On the other axis, at least one (here one) threshold may be set for the values identified by the sample rhythm data, namely threshold B1 (1 hour). This can form four mutually different sleep categories (here, sleep categories 10, 11, 20, 21) divided in terms of sleep quality and sleep rhythm by threshold A1 and threshold B1. In this example, sleep category 10 is the category with the highest probability of disturbed sleep, sleep category 21 is the category with the lowest probability of disturbed sleep, It can be said that sleep categories 11 and 20 are categories positioned between sleep categories 10 and 21 . Sleep category data may be data that identifies any of these four sleep categories.

FIG. 24 is a diagram showing another example of sleep category data used in communication system 1500 shown in FIG. In the example shown in FIG. 24, on the one axis, for example, two thresholds can be set for the score identified by the sample property data: threshold A1 (6 points) and threshold A2 (8 points). . On the other axis, for example, two thresholds can be set for the values identified by the sample rhythm data: threshold B1 (1 hour) and threshold B2 (3 hours). As a result, nine mutually different sleep categories (here, sleep categories 100, 101, 102, 200, 201) divided from the viewpoint of sleep quality and sleep rhythm by thresholds A1, A2 and thresholds B1, B2 , 202, 300, 301, 302) can be formed. In this example, sleep category 300 is the category most likely to disrupt sleep, sleep category 202 is the category most likely to disrupt sleep, It can be said that the remaining sleep categories are those that fall between sleep categories 300 and 202 . Sleep category data may be data identifying any of these nine sleep categories.

The examples shown in FIGS. 23 and 24 are merely examples for convenience, and any number of thresholds are set for the score identified by the sample property data, and the value identified by the sample rhythm data Any number of thresholds may be set for . By setting more thresholds for each of the sample quality data and the sample rhythm data, it is possible to generate more sleep categories that are different from each other, divided in terms of sleep quality and sleep rhythm. It is possible.

An information processing device, for example, a server device 1510A (and/or a terminal device 1520) managed by a company that operates an estimation service or the like, collects such sample behavior data and sample sleep state data for each sample user. In combination, a set of teacher data can be generated and stored for the sample user. In order to achieve this, for example, the information processing device extracts sample behavior data to which identification data for identifying a certain sample user is attached from among the plurality of acquired sample behavior data, Sample sleep state data attached with identification data for identifying the same sample user is extracted from the sleep state data, and combined to generate and store a set of teacher data for this sample user. be able to. By executing similar processing for each of a plurality of sample users, the information processing device can generate and store a plurality of sets of teacher data.

(2) Step 1002
Returning to FIG. 17, next, in ST1002, server device 1510A (and/or terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc. The estimated model can be generated by making the learning model learn by machine learning using the teacher data. Machine learning, as used herein, may include, but is not limited to, any well-known machine learning, such as gradient boosting trees (eg, LightGBM), neural networks, or linear regression.

The information processing apparatus can perform learning by sequentially inputting multiple sets of teacher data to a learning model. Here, each set of teacher data can include the sample behavior data of the sample user corresponding to the set and the sample sleep state data of the sample user corresponding to the set, as described above. The information processing apparatus can use the sample behavior data included in each set of teacher data as an explanatory variable, and can use the sample sleep state data included in that set of teacher data as an objective variable.

(2A) Example of machine learning In the case of using a gradient boosting tree, for example, when sample behavior data (explanatory variables) included in a certain set of teacher data is input to the learning data, the information processing device generates a sample sleep state A decision tree can be created and added to the learning model to improve the objective function calculated from the data (objective variable) and the estimated values (ie, the values output from the learning model). The information processing device can repeat this process for the number of decision trees defined by the hyperparameters. For the second and subsequent decision trees, learning can be performed on the difference between the objective variable and the estimated value from the decision trees created so far. In this way, the information processing device can determine the weights of the branches and leaves of each decision tree. That is, in the gradient boosting tree, each decision tree uses the explanatory variables of the teacher data as input data, and the values of the leaf nodes reached by following each branch of the decision tree as output data (objective variable). Furthermore, the method of each branch of the decision tree and the value of each leaf node can be optimized so that the error between this data (target variable) and the target variable of the teacher data is small.

In the case where the information processing device uses, for example, a neural network including an input layer, an intermediate layer, and an output layer, the information processing device inputs the explanatory variables of the teacher data to the input layer and outputs them from the output layer. It is possible to optimize (learn) each parameter (weight) included in the neural network using the error backpropagation method so as to reduce the error between the data (target variable) and the target variable in the teacher data. can.

In the case of using linear regression, the information processing device can learn the parameters of the linear regression model so as to improve the error between the objective variable and the estimated value. This linear regression model is divided into an input layer and an output layer. The least squares method can be used to optimize the parameters included in the linear regression model so that the error with the objective variable is small.

(2B) In an example of the first plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the first plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the first plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of teacher data can include sample disposition data of the sample user corresponding to the set and sample rhythm data of the sample user.

In this case, the information processing device can perform preprocessing on each set of teacher data included in the first plurality of sets of teacher data. Specifically, the information processing device performs the processing described above with reference to FIGS. can be pre-processed to generate sample sleep category data. As a result, the information processing device provides the first plurality of sets of preprocessed teacher data (each set of teacher data includes the sample behavior data of the sample user corresponding to the set and the sample sleep category data of the sample user. , ) can be input into the learning model to generate (update) the estimation model. As will be described later, such an estimation model responds to inputting, as an explanatory variable, target action data that identifies a history of actions performed by the target user, and identifies the sleep category to which the target user belongs. At least the sleep category data can be output.

(2C) In an example of the second plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the second plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the second plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of teacher data can include sample sleep category data for the sample users corresponding to that set.

In this case, the information processing device includes a second plurality of sets of teacher data (each set of teacher data includes sample behavior data of the sample user corresponding to the set and sample sleep category data of the sample user). can be input to the learning model to generate (update) the estimation model. As will be described later, such an estimation model responds to inputting, as an explanatory variable, target action data that identifies a history of actions performed by the target user, and identifies the sleep category to which the target user belongs. At least the sleep category data can be output.

(2D) In an example of the third plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the third plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the third plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of teacher data can include sample disposition data of the sample user corresponding to the set and sample rhythm data of the sample user.

In this case, the information processing device generates a third plurality of sets of teacher data (each set of teacher data includes sample behavior data of the sample user corresponding to the set, sample behavior data of the sample user, sample property data of the sample user, and sample rhythm data) can be input into the learning model to generate (update) the estimation model. Such an inference model, as described below, responds to inputting, as an explanatory variable, target behavior data that identifies a history of behavior performed by the target user, the target At least the property data and the target rhythm data identifying the target user's sleep rhythm can be output.

(2E) In an example of a fourth plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the fourth plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the fourth plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of training data can include sample disposition data of the sample users corresponding to that set.

In this case, the information processing device generates a fourth set of teacher data (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample property data of the sample user). , can be input to the learning model to generate (update) the estimation model. Such an inference model, as described below, responds to inputting, as an explanatory variable, target behavior data that identifies a history of behavior performed by the target user, the target At least the property data can be output.

(2F) In an example of a fifth plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the fifth plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the fifth plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of teacher data can include the sample rhythm data of the sample user corresponding to that set.

In this case, the information processing device generates a fifth plurality of sets of teacher data (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample rhythm data of the sample user). , can be input to the learning model to generate (update) the estimation model. As will be described later, such an estimation model responds to inputting, as an explanatory variable, target action data that identifies a history of actions performed by the target user, to identify the sleep rhythm of the target user. At least rhythm data can be output.

(2G) In an example of the sixth plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the fifth plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the sixth plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample sleep state data of the sample user corresponding to the set. . Here, the sample sleep state data in each set of teacher data can include sample time data of sample users corresponding to the set.

In this case, the information processing device generates a sixth plurality of sets of teacher data (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample time data of the sample user). , can be input to the learning model to generate (update) the estimation model. Such an estimation model, as will be described later, responds to inputting, as an explanatory variable, target behavior data that identifies a history of behaviors performed by the target user. Data can at least be output.

Generation (updating) of such an estimation model can be performed instead of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or managed by the company that operates the estimation service or the like. may be performed by other server devices 1510 (and/or terminal devices 1520) managed by other companies, as well as by server device 1510A (and/or terminal device 1520). The estimation model generated (updated) by the other server device 1510 (terminal device 1520) can be transmitted to any server device 1510 including the server device 1510A, or any server device including the server device 1510A. It can be accessed and used by server device 1510 .

It should be noted that the names "first" to "sixth" attached immediately before "multiple sets of teacher data" are referred to as "multiple sets of teacher data" with a certain name among these names. , and "multiple sets of teacher data" with different names among these names, and are used for convenience, and are not intended to limit any order or content.

(3) Step 1004
Referring to FIG. 17 again, next, in ST1004, the server device 1510A (and/or the terminal device 1520) which is an information processing device, for example, managed by a company that operates an estimation service or the like, executes a specific application, for example. From the terminal device 1520 (or another server device 1510) that has accessed this server device 1510A by executing an application (such as an application for receiving provision of an estimation service, etc.) and/or this server device by executing a browser From the terminal device 1520 that has accessed 1510A, it is possible to acquire target action data that identifies the history of actions performed by the target user to be inferred. This target action data may be data identifying at least one (one or more) actions performed by the target user.

Acquisition of such target behavior data can be performed in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by a server managed by the company that operates the estimation service or the like. Along with device 1510A (and/or terminal 1520), it may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

(4) Step 1006
Next, in ST1006, the server device 1510A (and/or the terminal device 1520), which is an information processing device, for example, managed by a company that operates an estimation service or the like, processes the target behavior data acquired in ST1004 and ST1002 The estimated model generated in can be used to obtain target sleep state data identifying sleep states of the target user. Specifically, the server device 1510A can cause the estimation model to output the target sleep state data as the objective variable by inputting the target behavior data as the explanatory variable into the estimation model.

The targeted sleep state data can include at least one of the following data.
・Data that identifies the sleep characteristics of the target user (“object attribute data”)
・Data identifying the sleep rhythm of the target user (“target rhythm data”)
・Data identifying the sleep time of the target user ("target time data")
・Data identifying the sleep category to which the target user (sleeping) belongs ("target sleep category data")

Which data the estimation model outputs as the target sleep state data may depend on which data was used as the sample sleep state data when generating (updating) the estimation model.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the first plurality of sets of teacher data is used (the above "6.4 (2) (2B)" section), The estimation model can at least output the target sleep category data as the target sleep state data.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the second plurality of sets of teacher data is used (section "6.4 (2) (2C)" above), The estimation model can at least output the target sleep category data as the target sleep state data.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the third set of teacher data is used (the above "6.4 (2) (2D)" section), The estimation model can output at least the target property data and the target rhythm data as the target sleep state data.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the fourth set of teacher data is used (above "6.4 (2) (2E)" section), The estimation model can at least output the target property data as the target sleep state data.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the fifth set of teacher data is used (above "6.4 (2) (2F)" section), The estimation model can at least output the target rhythm data as the target sleep state data.

For example, when the estimation model is generated (updated), as the sample sleep state data, when the sixth plurality of sets of teacher data is used (the above "6.4 (2) (2G)" section), The estimation model can at least output the target time data as the target sleep state data.

For example, as sample sleep state data, an estimation model is generated (updated) using the first plurality of sets of teacher data (Section 6.4 (2) (2B) above), and then sample sleep When the estimation model is generated (updated) using the sixth plurality of sets of teacher data as the state data (section “6.4 (2) (2G)” above), the estimation model is At least the target sleep category data and the target time data can be output as the sleep state data. Furthermore, after this, as sample sleep state data, the estimation model was generated (updated) using the fifth set of teacher data (section "6.4 (2) (2F)" above). , the estimation model can output at least the target sleep category data, the target time data and the target rhythm data as the target sleep state data.

As a method of using the estimation model, it is possible to use any of the methods exemplified below.
- The server device 1510A acquires and holds the estimation model generated in ST1004, inputs the target behavior data into this estimation model, and acquires the target sleep state data output from this estimation model.
- The server device 1510A transmits target behavior data via the communication network 2 to an external device (any other server device 1510 or any terminal device 1520) that acquires and holds the estimation model generated in ST1004. to output the target sleep state data to the estimation model that has input the target behavior data as an explanatory variable. After that, the server device 1510A receives the target sleep state data output by this estimation model from the external device via the communication network 2 .

Acquisition of such target sleep state data is performed instead of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or managed by the company that operates the estimation service or the like. Along with server device 1510A (and/or terminal device 1520), it may be performed by other server device 1510 (and/or terminal device 1520) managed by other companies.

(5) Step 1008
Next, in ST1008, an information processing device, for example, a server device 1510A (and/or a terminal device 1520) managed by a company that operates an estimation service, uses the target sleep state data acquired in ST1006, At least one target product/one target service corresponding to the target sleep state data can be determined. Such targeted goods/services may be appropriate for the current (estimated) sleep state of the targeted user identified by the targeted sleep state data.

Specifically, in one example, the server device 1510A can first prepare and store a search table that associates and stores the target sleep state data with a plurality of products and/or a plurality of services. Next, server device 1510A inputs the target sleep state data acquired in ST1006 into the search table, thereby identifying at least one target product and/or at least one target service from this search table ("proposal data”) can be obtained.

FIG. 25 is a diagram conceptually showing an example of a search table used in communication system 1500 shown in FIG. The search table exemplified in FIG. 25 is associated with each data (target property data, target rhythm data, target time data, and target sleep category data) that can be included in the target sleep state data, and at least one data to be proposed. stores one target product and/or at least one target service.

In this example, for simplification of explanation, at least one level to be proposed is associated with each of three mutually different levels indicated by each of the target property data, the target rhythm data, the target time data, and the target sleep category data. One target product and/or at least one target service may be assigned. Taking the target property data as an example, the target property data indicates class V1 (a class in which the target user's sleep property is good), class V2 (a class in which the target user's sleep property is standard), and class V3 ( At least one target product and/or at least one target service to be proposed may be associated with each of the target user's poor sleep quality classes.

Classes W1-W3 may also indicate, in that order, the subject user's sleep rhythm is "good", "normal", and "bad". Classes X1-X3 may also indicate, in that order, the subject user's sleep time is "long", "normal", and "short".

Classes Y1-Y3 may also indicate, in this order, the sleep category to which the target user's (sleep) belongs: "Good", "Normal", and "Poor". In one example, class Y1 corresponds to "sleep category 21" illustrated in FIG. 23, class Y2 corresponds to "sleep category 20" or "sleep category 11" in the same figure, and class Y3 corresponds to " It can also be considered that it corresponds to sleep category 10.

All of the products P1 to P11 illustrated in FIG. 25 are arbitrary products suitable for improving the sleep state of the user, and the services S1 to S8 illustrated in FIG. Indicates any service suitable for improving sleep conditions. These commodities can include, but are not limited to, food, household items, medicines, exercise equipment, electrical appliances, and the like. These services may also include, but are not limited to, hospitals, clinics, gyms, and the like.

For example, when the target property data in the target sleep state data indicates class V3, server device 1510A can determine product P4 and service S1 as the target product and target service, respectively. For example, when the target sleep state data includes target time data (indicating class X2) and target rhythm data (indicating class W3), server device 1510A treats product P8 as the target product and services S1 and S3. , S4 can be determined as the target service.

For each data included in the target sleep state data, the total number of classes used, the total number of products and/or services assigned to each class, and the like can be arbitrarily determined.

In another example, server device 1510A uses another inference model generated by performing supervised learning (machine learning) to determine at least one target product and/or at least one target service. is also possible. The other estimation model uses data indicating each class of each data included in the target sleep state data as an explanatory variable, and data indicating at least one target product and/or at least one target service as an objective variable, It can be generated (updated) by performing supervised learning.

Machine learning for training the other estimation model may be any well-known machine learning, such as gradient boosted trees (e.g., LightGBM), neural networks, linear regression, etc., as described in section "2A" above. can include, but are not limited to,

By inputting the target sleep state data into the separate estimation model, the server device 1510A provides data identifying at least one target product and/or at least one target service ("suggestion data") from the separate estimation model. ) can be output.

Any one of the methods exemplified below can be used as a method of using the other estimation model.
The server device 1510A acquires and holds the generated another estimation model, inputs the target sleep state data to the another estimation model, and acquires the proposal data output from the another estimation model. .
· The server device 1510A provides an external device (any other server device 1510 or any terminal device 1520) that acquires and holds the generated another estimation model via the communication network 2. By transmitting the state data, the other estimation model to which the target sleep state data is input as an explanatory variable is caused to output proposal data. After that, the server device 1510A receives the proposal data output by the other estimation model from the external device via the communication network 2 .

Acquisition of such proposal data can be performed in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by the server device managed by the company that operates the estimation service or the like. 1510A (and/or terminal 1520) may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

(6) Step 1010
Referring to FIG. 17 again, next, the server device 1510A (and/or the terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc., converts the proposal data acquired in ST1008. , can be provided (transmitted) via the communication network 2 to at least one terminal device 1520 (which may be another server device 1510) managed by a corporate user and/or an individual user. "Individual users" as used herein may include "target users."

Proposal data can be provided by corporate users and/or individual users by any means including web pages, e-mails, chats, electronic files, recording media (USB memory, CD-ROM, etc.), and/or paper media. may be provided to at least one terminal device 1520 (which may be another server device 1510) managed by .

Such proposal data is provided in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by the server device managed by the company that operates the estimation service or the like. 1510A (and/or terminal 1520) may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

6.5 Modifications In the various examples described above, the cases where the sample property data, the sample rhythm data, the sample time data, and the sample sleep category data are used as objective variables to generate and use an estimation model have been described. In another example, at least one (up to three) of sample disposition data, sample rhythm data, sample time data, and sample sleep category data are used as explanatory variables rather than objective variables to generate and use an estimation model. It is also possible to

Further, in the various examples described above, the executing entity of the various steps (eg, ST1000 to ST1010) illustrated in FIG. 17 is mainly the server device 1510A (and/or The case of the terminal device 1520) has been described. However, the various steps illustrated in FIG. 17 can also be shared and executed by arbitrary multiple server devices 1510 and/or multiple terminal devices 1520 . Here, any of the plurality of server devices 1510 and/or the plurality of terminal devices 1520 may include, for example, a server device 1510A (and/or the terminal device 1520) managed by a company that operates an estimation service or the like. In another example, server device 1510A (and/or terminal device 1520) managed by a company that operates an estimation service or the like is not included.

In "6.4 (4) step 1006" above, definitions of each of the target property data, target rhythm data, target time data, and target sleep category data that can be included in the target sleep state data are shown. In another embodiment, the target sleep state data can include at least one of the data shown below instead of the data shown in "6.4 (4) Step 1006" above.
・Data that “uniquely” identifies the sleep characteristics of the target user (“target attribute data”)
・Data that “uniquely” identifies the sleep rhythm of the target user (“target rhythm data”)
・Data that “uniquely” identifies the sleep time of the target user (“target time data”)
・Data that “uniquely” identifies the sleep category to which the target user (sleeping) belongs (“target sleep category data”)

Here, "unique identification" by the "target property data" means that the sleep property of the target user can be identified without using other data other than the "target property data". means. That is, when a plurality of data cooperate to identify the sleep properties of the target user, these multiple data do not correspond to data that "uniquely identifies" the sleep properties of the target user. . For example, the first data includes data indicating that the target user's sleep quality is "poor" and a flag "0", and the second data indicates that the target user's sleep quality is "normal". and the flag "0", and the third data includes data and the flag "1" indicating that the sleep quality of the target user is "high". In this case, the first data and the second data alone cannot determine the sleep behavior of the target user. Only by checking the third data with flag "1" can it be determined that the sleep quality of the target user is "high". Therefore, none of these data is data that "uniquely" identifies the sleep quality of the intended user.

The term "uniquely identify" included in the above definitions of each of "target rhythm data", "target time data" and "target sleep category data" can be similarly considered.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Targeted sleep state data can be obtained that identifies the user's current (predicted) state of sleep. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target sleep state data acquired in this way is provided to the target user. can be suggested to any user, including the target user, to improve their sleep.

Thereby, a computer program, an information processing apparatus, and a method for estimating a sleep state of a target user based on a history of at least one action performed by the target user, and having at least partially improved performance. can provide a method.

6.6 Actions Performed by Communication System 1500 to Provide Menopausal Status Data Regarding Menopausal Symptom Status Next performed by communication system 1500 described above to provide menopausal status data about menopausal symptom status A specific example of the operation will be further described with reference to FIG. FIG. 26 is a flow diagram illustrating another example of operations performed by the communication system 1500 shown in FIG.

(1) Step 1100
First, in ST1100, an information processing device, for example, a server device 1510A managed by a company that operates an estimation service or the like acquires and stores multiple sets of teacher data used to generate an estimation model. can be done.

Each set of teacher data constituting a plurality of sets of teacher data includes data identifying the history of actions performed by one sample user corresponding to this set ("sample action data") and and data identifying the menopausal symptom status of a sample user ("sample menopausal status data"). For example, the first set of teacher data includes sample behavior data identifying a history of behaviors performed by one sample user (Mr. V) corresponding to the first set, and identifying Mr. V's menopausal symptom status. sample menopausal status data, wherein the second set of teacher data includes sample behavior data identifying a history of behavior performed by one sample user (Mr. W) corresponding to the second set; and sample menopausal status data identifying Mr. W's menopausal symptom status.

(1A) Sample behavior data Each sample behavior data may be data that identifies at least one (one or more) behavior among a plurality of predetermined behaviors. Here, each of the plurality of predetermined actions may be actions that can at least partially have a positive or negative effect on the state of menopausal symptoms of the person who performed the action. Such predetermined behaviors can include, for example, but not limited to, those exemplified below.
(Examples of behaviors that can at least partially positively affect a person's menopausal symptom status)
・Performed aerobic exercise for 1 hour or more at a pace of 2 days or more per week.
・I got up at a certain time every day.
- Set aside time to relax every week.
- Purchased 10 or more cans of beer (this behavior can be identified by data including, for example, the product category to which the beer belongs and its purchase quantity).
(Examples of behaviors that can at least partially adversely affect a person's menopausal symptom status)
- Did aerobic exercise for 1 hour or more at a pace of 1 day or less per week.
-Woke up at different times every day.
・I worked without taking a day off a week.
- Purchased 5 or more bottles of green juice (this behavior can be identified by data including, for example, the product category to which the green juice belongs and its purchase quantity).

Each sample behavior data may be data identifying each of at least one behavior among a plurality of predetermined behaviors exemplified in this way.

Also, in one example, since the product purchased by the user and/or the service used by the user may tend to have a positive or negative effect on the state of menopausal symptoms of the user, the plurality of predetermined behaviors can also include the product purchased by the sample user (or the category to which the product belongs), the service used by the sample user (or the category to which the service belongs). This type of action can be generated using, for example, POS (Point Of Sales) data held by retail stores (drug stores, supermarkets, convenience stores, etc.).

In one example, each of a plurality of predetermined behaviors is assigned unique identification data (such as letters, numbers, or combinations thereof), and each sample behavior data may include such identification data. can. For example, when using POS data, each of the plurality of sample behavior data is a JICFS classification (classification code), product category, purchase amount, purchase quantity, purchase frequency, and/or corresponding to the purchased product. , date of purchase, etc., without limitation.

For example, a server device 1510A managed by a company that operates an estimation service or the like, which is an information processing device, can acquire such sample behavior data by, for example, at least one of the methods exemplified below. can.
- The information processing device, via the communication network 2, at least one other server device 1510 (for example, another server device 1510 that stores POS data installed in a drugstore or the like, or can access this POS data) POS data is received from another server device 1510).
- The information processing device sends an e-mail or a web page that describes the answers to a predetermined questionnaire (questionnaire about actions taken by the sample users) via the communication network 2 to at least one terminal device 1520 and/or at least one Received from server device 1510 .
- The information processing device receives, via the communication network 2, the content of the responses from at least one server device 1510 that has collected responses from a plurality of users to the predetermined questionnaire.
- The information processing device receives the contents of the answers from a plurality of users to the predetermined questionnaire via a recording medium (USB memory, DVD-ROM, etc.).

(1B) Sample menopausal state data The sample menopausal state data included in each set of teacher data can include at least one of the data exemplified below, without being limited thereto.
- data identifying the overall state of one sample user corresponding to the set ("sample overall state data");
- data identifying the state of the first symptom of one sample user corresponding to the set ("sample first state data");
- data identifying the second symptom state of one sample user corresponding to the set ("sample second state data");

First, the sample overall condition data can be data identifying whether the sample user's menopausal condition is good (eg, how good) or bad (eg, how bad). In one example, the sample overall condition data may be data identifying a score that comprehensively assesses the condition of the sample user's menopausal symptoms. In this case, the sample user's menopausal symptom status may be, for example, data identifying a score calculated by any of the methods exemplified below.

・Score calculated by the method described in the SMI based on the sample user's answers to the questions described in the Simplified Menopause Index (SMI) (see FIG. 27), or this score is calculated by any method Corrected score obtained by (addition, subtraction, multiplication, division, etc.) Calculated by the method described in the menopause rating scale based on the sample user's answers to the questions described in the menopause rating scale or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, etc.) Based on the contents of the sample user's answers to the questions described in the Kupperman index , a score calculated by the method described in the Kupperman climacteric index, or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, etc.) Menopausal disorder evaluation (Green Climatic Scale) described Score calculated by the method described in the menopausal disorder evaluation based on the content of the sample user's answers to the questionnaire items, or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, etc.)・ Based on the answers of sample users to the questions described in PSST (Premenstrual Symptoms Screening Tool), the score calculated by the method described in PSST, or this score is calculated by any method (addition, subtraction, multiplication, etc. ) based on the sample user's answers to the questions described in the WHQ (Women's Health Questionnaire), the score calculated by the method described in the WHQ, or this score Corrected score obtained by calculating (addition, subtraction, multiplication, division, etc.) by any method ・Calculated by the method described in the VAS (Visual Analogue Scale) based on the answers of sample users to the questions described in the VAS or a modified score obtained by calculating (addition, subtraction, multiplication, division, etc.) this score by an arbitrary method. Based on the score calculated by the method described in HFRDI A, or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, division, etc.) ・HFCS (Hot Flash Composite Score) based on the answers of sample users to the questions described in Score calculated by the method described in , or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, division, etc.) ・Questions described in MENQOL (Menopause-Specific Quality Of Life) Score calculated by the method described in MENQOL based on the answers of the sample users, or a modified score obtained by calculating this score by any method (addition, subtraction, multiplication, division, etc.) ・ Japanese women's A score calculated by the method described in the menopausal symptom evaluation table for Japanese women based on the sample user's answers to the questions described in the menopausal symptom evaluation table, or this score is calculated by any method ( Modified score obtained by adding, subtracting, multiplying, multiplying, etc.) ・Calculated by the method described in any other questionnaire or scale based on the content of the sample user's answers to the questions described in that questionnaire or scale. score, or a modified score obtained by computing this score by any method

Second, the sample first condition data identifies whether the sample user is in good (e.g., how good) or bad (e.g., how bad) condition of one of the plurality of menopausal symptoms. can be data. In one example, the sample first condition data may be data identifying a score evaluating the condition of the one symptom for the sample user.

Taking SMI as an example, the plurality of climacteric symptoms may be a plurality of climacteric symptoms described in SMI, such as "feeling hot," "easily sweating," and "easily getting cold in the waist and limbs." In addition to SMI, the plurality of climacteric symptoms may be a plurality of climacteric symptoms described in any of the indices listed above. Also, the one symptom may be any one symptom selected from such a plurality of menopausal symptoms.

Taking the SMI as an example, the sample first state data can be extracted, in one example, from the content of the sample user's answers to the questions described in the SMI (see FIG. 27). In this case, the sample first condition data is, for example, data identifying a certain symptom condition such as "shortness of breath and palpitations", and for example, a score of 0, 4, 8 or 12 (see FIG. 27). It can be identifying data. In another example, sample first state data may be obtained from the sample users via e-mail, web pages, or the like.

Third, the sample second condition data identifies whether the condition of another one of the plurality of menopausal symptoms for the sample user is good (eg, how good) or bad (eg, how bad). can be data. In one example, the sample second condition data may be data identifying a score evaluating the condition of the another one of the sample users.

Taking SMI as an example, the plurality of climacteric symptoms may be a plurality of climacteric symptoms described in SMI, such as "feeling hot," "easily sweating," and "easily getting cold in the waist and limbs." In addition to SMI, the plurality of climacteric symptoms may be a plurality of climacteric symptoms described in any of the indices listed above. Also, the another symptom may be any one symptom selected from such a plurality of menopausal symptoms, but may be a different symptom from the certain one symptom.

Taking the SMI as an example, the sample second state data can be extracted, for example, from the content of the sample user's answers to the questions written in the SMI (see FIG. 27). In this case, the sample second condition data is, for example, data identifying another symptom condition of "easily tired", for example, data identifying a score of 0, 2, 4, or 7 (see FIG. 27). can be In another example, sample second state data may be obtained from the sample user via e-mail, web page, or the like.

For example, a server device 1510A (and/or a terminal device 1520) managed by a company that operates an estimation service or the like, which is an information processing device, collects such sample behavior data and sample menopausal state data for each sample user. In combination, a set of teacher data can be generated and stored for the sample user. In order to achieve this, for example, the information processing device extracts sample behavior data to which identification data for identifying a certain sample user is attached from among the plurality of acquired sample behavior data, Sample menopausal state data to which identification data for identifying the same sample user is extracted from the menopausal state data, and combined to generate and store a set of teacher data for this sample user. be able to. By executing similar processing for each of a plurality of sample users, the information processing device can generate and store a plurality of sets of teacher data.

(2) Step 1102
Returning to FIG. 17, next, in ST1102, server device 1510A (and/or terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc. The estimated model can be generated by making the learning model learn by machine learning using the teacher data. Machine learning, as used herein, may include, but is not limited to, any well-known machine learning, such as gradient boosting trees (eg, LightGBM), neural networks, or linear regression.

(2A) In an example of the first plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the first plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the first plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample menopausal state data of the sample user corresponding to the set. . Here, the sample menopausal status data in each set of teacher data can include sample overall status data for the sample users corresponding to the set.

In this case, the information processing device provides the first plurality of sets of teacher data (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample general condition data of the sample user). can be input to the learning model to generate (update) the estimation model. Such an inference model identifies the overall state of the target user in response to inputting, as an explanatory variable, target behavior data identifying the history of actions performed by the target user, as described below. At least the target general condition data can be output.

In one embodiment, the sample menopausal state data in each set of teacher data identifies the overall state of menopausal symptoms of one sample user corresponding to the set. think of.

In this case, the information processing device can perform preprocessing on each set of teacher data included in the first plurality of sets of teacher data. Specifically, for each set of teacher data, the information processing device uses at least one threshold value to determine the score identified by the sample general state data corresponding to this set as one of a plurality of categories. Pre-processing can be performed to convert to categories. For example, when using one threshold (first threshold), the information processing device determines whether the score identified by the sample general state data corresponding to each set is greater than or equal to the first threshold. It can be converted into one of two categories (“High” and “Low”) depending on whether it is less than. Alternatively, for example, when the information processing device uses two thresholds (a first threshold and a second threshold smaller than the first threshold), the score identified by the sample comprehensive state data corresponding to each set is , three categories (“High”, “Middle” and “Low ) can be converted into any of the following categories:

In this way, the information processing device corresponds to the general state of menopausal symptoms of one sample user corresponding to each set of teacher data included in the first plurality of sets of teacher data. The pre-processing of transforming the sample aggregate condition data identifying the "score" into sample aggregate condition data identifying the "category corresponding to" the aggregate condition of one sample user's menopausal symptoms corresponding to this set. can be executed.

As a result, the information processing device stores the first plurality of sets of teacher data after preprocessing (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample general state data of the sample user). , ) can be input into the learning model to generate (update) the estimation model.

Such an inference model, as described below, responds to inputting, as an explanatory variable, target behavior data that identifies a history of behaviors performed by the target user. At least target aggregate condition data that identifies the category corresponding to the can be output.

(2B) In an example of the second plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the second plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the second plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample menopausal state data of the sample user corresponding to the set. . Here, the sample menopausal state data in each set of teacher data can include sample first state data of the sample user corresponding to the set.

In this case, the information processing device includes the second plurality of sets of teacher data (each set of teacher data includes the sample behavior data of the sample user corresponding to the set and the sample first state data of the sample user). ) can be input to the learning model to generate (update) the estimation model. As will be described later, such an inference model identifies the state of the target user's first symptom in response to inputting, as an explanatory variable, target behavior data that identifies a history of actions performed by the target user. It becomes possible to output at least the target first state data.

In one embodiment, the sample first condition data in each set of teacher data identifies the "score corresponding to" the state of the first symptom of one sample user corresponding to the set. think.

In this case, the information processing device can perform preprocessing on each set of teacher data included in the second plurality of sets of teacher data. Specifically, for each set of teacher data, the information processing device uses at least one threshold value to determine the score identified by the sample first state data corresponding to this set in one of a plurality of categories. can be pre-processed to convert to categories of For example, when using one threshold (first threshold), the information processing device determines whether the score identified by the sample first state data corresponding to each set is equal to or greater than the first threshold or the first threshold. Depending on whether it is below the threshold, it can be converted into one of two categories (“High” and “Low”). Alternatively, for example, when the information processing device uses two thresholds (a first threshold and a second threshold smaller than the first threshold), the score identified by the sample first state data corresponding to each set is divided into three categories ("High", "Middle" and " Low").

In this way, the information processing device calculates the "score corresponding to" the state of the first symptom of one sample user corresponding to each set of teacher data included in the second plurality of sets of teacher data. into sample first state data that identifies the "category corresponding to" the state of the first symptom of the one sample user corresponding to this set be able to.

As a result, the information processing device stores the second plurality of sets of teacher data after preprocessing (each set of teacher data consists of sample behavior data of a sample user corresponding to the set and sample first state data of the sample user). ) can be input into the learning model to generate (update) the estimation model.

As will be described later, such an estimation model corresponds to the state of the target user's first symptom in response to inputting, as an explanatory variable, target action data that identifies a history of actions performed by the target user. At least the subject first state data identifying the category to be played can be output.

Furthermore, the information processing apparatus can perform further preprocessing (“preprocessing B” for convenience) in addition to the preprocessing described above (“preprocessing A” for convenience). Specifically, first, the information processing apparatus can execute "preprocessing B" for extracting another plurality of sets of teacher data from the second plurality of sets of teacher data. Each set of teacher data included in the other plurality of sets of teacher data has sample first state data identifying scores equal to or greater than a threshold.

For example, when the first symptom is the symptom of "prone to sweating" in SMI (see FIG. 27), each set of teacher data included in the separate multiple sets of teacher data has a score equal to or higher than a threshold of 5 ( 10 or 6). That is, each set of teacher data included in the separate multiple sets of teacher data includes teacher data only for sample users who have a stronger first symptom (here, "prone to sweating").

Furthermore, the information processing device can use the other plural sets of teacher data thus extracted as a "new" second plural set of teacher data. Thereafter, the information processing device can perform the above-described preprocessing A on the "new" second plurality of sets of teacher data. In this pre-processing A, the at least one threshold to be used can be set to a larger value than in the example above where pre-processing B is not performed. Next, the information processing apparatus can input the "new" second plurality of sets of teacher data after preprocessing A into the learning model to generate (update) the estimation model.

Such a "new" second set of teacher data includes sample behavior data and sample first condition data only for sample users with stronger first symptoms. Therefore, the estimation model trained using such a "new" second plurality of sets of teacher data is able to identify target users who have performed behaviors that may lead to strong first symptoms. can be more easily estimated as a user having .

(2C) In an example of the third plurality of sets of teacher data included in the plurality of sets of teacher data , the information processing device uses the third plurality of sets of teacher data included in the plurality of sets of teacher data to generate an estimation model. It can be generated (updated). Each set of teacher data included in the third plurality of sets of teacher data can include sample behavior data of the sample user corresponding to the set and sample menopausal state data of the sample user corresponding to the set. . Here, the sample menopausal state data in each set of teacher data can include sample second state data of the sample user corresponding to the set.

In this case, the information processing device includes a third set of teacher data (each set of teacher data includes sample behavior data of a sample user corresponding to the set and sample second state data of the sample user). ) can be input to the learning model to generate (update) the estimation model. Such an estimation model identifies the state of the second symptom of the target user in response to inputting, as an explanatory variable, target behavior data that identifies a history of actions performed by the target user, as described below. It becomes possible to output at least the target second state data.

In one embodiment, the sample second condition data in each set of teacher data identifies the "score corresponding to" the condition of the second symptom of one sample user corresponding to the set. think.

In this case, the information processing device can perform preprocessing on each set of teacher data included in the third plurality of sets of teacher data. Specifically, for each set of teacher data, the information processing device uses at least one threshold value to determine the score identified by the sample second state data corresponding to this set in one of a plurality of categories. can be pre-processed to convert to categories of For example, when using one threshold value (first threshold value), the information processing device determines whether the score identified by the sample second state data corresponding to each set is equal to or greater than the first threshold value. Depending on whether it is below the threshold, it can be converted into one of two categories (“High” and “Low”). Alternatively, for example, when the information processing device uses two thresholds (a first threshold and a second threshold smaller than the first threshold), the score identified by the sample second state data corresponding to each set is divided into three categories ("High", "Middle" and " Low").

In this way, the information processing device calculates the "score corresponding to" the state of the second symptom of one sample user corresponding to each set of teacher data included in the third plurality of sets of teacher data. into sample second state data that identifies the "category corresponding to" the state of the second symptom of the one sample user corresponding to this set be able to.

As a result, the information processing device generates the third plural sets of teacher data after preprocessing (each set of teacher data consists of the sample behavior data of the sample user corresponding to the set and the sample first state data of the sample user). ) can be input into the learning model to generate (update) the estimation model.

As will be described later, such an estimation model corresponds to the state of the target user's second symptom in response to inputting, as an explanatory variable, target behavior data that identifies a history of actions performed by the target user. At least the target second state data identifying the category to be played can be output.

Furthermore, the information processing apparatus can perform further preprocessing (“preprocessing B” for convenience) in addition to the preprocessing described above (“preprocessing A” for convenience). Specifically, first, the information processing device can execute "preprocessing B" for extracting another plurality of sets of teacher data from the third plurality of sets of teacher data. Each set of teacher data included in the other plurality of sets of teacher data has sample second state data identifying scores equal to or greater than a threshold.

For example, when the second symptom is the symptom of "easily tired" in SMI (see FIG. 27), each set of teacher data included in the separate multiple sets of teacher data has a score above the threshold of 3 (7 or 4) may be sample second state data that identifies 4). That is, each set of teacher data included in the separate multiple sets of teacher data includes teacher data only for sample users who have a stronger second symptom (here, "easily tired").

Further, the information processing device can use the other plural sets of teacher data thus extracted as a "new" third plural set of teacher data. After this, the information processing device can perform the above-described preprocessing A on the "new" third set of teacher data. In this pre-processing A, the at least one threshold to be used can be set to a larger value than in the example above where pre-processing B is not performed. Next, the information processing apparatus can input the "new" third plurality of sets of teacher data after preprocessing A into the learning model to generate (update) the estimation model.

Such a "new" third set of teacher data includes sample behavior data and sample second condition data only for sample users with stronger second symptoms. Therefore, the estimation model that has been trained using such a "new" third plurality of sets of teacher data can identify target users who have performed behaviors that may lead to strong second symptoms. can be estimated more easily as a user having .

Generation (updating) of the estimation model as described above is performed instead of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or managed by the company that operates the estimation service or the like. can be performed by other server devices 1510 (and/or terminal devices 1520) managed by other companies, as well as by server device 1510A (and/or terminal device 1520) managed by other companies. The estimation model generated (updated) by the other server device 1510 (terminal device 1520) can be transmitted to any server device 1510 including the server device 1510A, or any server device including the server device 1510A. It can be accessed and used by server device 1510 .

Note that the names "first" to "third" attached immediately before "multiple sets of teaching data" are referred to as "multiple sets of teaching data" with a certain name among these names. , and "multiple sets of teacher data" with different names among these names, and are used for convenience, and are not intended to limit any order or content.

(3) Step 1104
Referring to FIG. 26 again, next, in ST1104, server device 1510A (and/or terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc. From the terminal device 1520 (or another server device 1510) that has accessed this server device 1510A by executing an application (such as an application for receiving provision of an estimation service, etc.) and/or this server device by executing a browser From the terminal device 1520 that has accessed 1510A, it is possible to acquire target action data that identifies the history of actions performed by the target user to be inferred. This target action data may be data identifying at least one (one or more) actions performed by the target user.

Acquisition of such target behavior data can be performed in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by a server managed by the company that operates the estimation service or the like. Along with device 1510A (and/or terminal 1520), it may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

(4) Step 1106
Next, in ST1106, the server device 1510A (and/or the terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc., processes the target behavior data acquired in ST1104 and ST1102 The estimation model generated in can be used to obtain target menopausal condition data that identifies the state of the target user's menopausal symptoms. Specifically, the server device 1510A can input the target behavior data as explanatory variables into the estimation model, thereby outputting the target menopausal state data as the objective variable from the estimation model.

The targeted menopausal status data can include at least one of the data set forth below.
・Data identifying the general condition of the target user's menopausal symptoms ("target general condition data")
・Data identifying the state of the first symptom of the target user (“target first state data”)
・Data identifying the state of the second symptom of the target user (“target second state data”)

Which data the estimation model outputs as the target menopausal state data may depend on which data was used as the sample menopausal state data when the estimation model was generated (updated).

For example, when the estimation model is generated (updated), as the sample menopausal state data, when the first plurality of sets of teacher data is used (above "6 (2) (2A)"), the estimation model can output at least the target general condition data as the target menopausal condition data.

For example, when the estimation model is generated (updated), as the sample menopausal state data, when the second plurality of sets of teacher data is used (above "6 (2) (2B)"), the estimation model can output at least the target first condition data as the target menopausal condition data.

For example, when the estimation model is generated (updated), as the sample menopausal state data, when the third plurality of sets of teacher data is used (above "6 (2) (2C)"), the estimation model can output at least the target second condition data as the target menopausal condition data.

For example, as sample menopausal state data, an estimation model is generated (updated) using the first plurality of sets of teacher data (item "6 (2) (2A)" above), and then sample menopausal state data As such, when the estimation model is generated (updated) using the second plurality of sets of teacher data (item "6 (2) (2B)" above), the estimation model is the target menopausal state data , the target general state data and the target first state data can be output at least. Furthermore, after this, when the estimation model is generated (updated) using the third plurality of sets of teacher data as the sample menopause state data (item "6 (2) (2C)" above), the estimation The model can output at least the target general condition data, the target first condition data, and the target second condition data as the target menopausal condition data.

As a method of using the estimation model, it is possible to use any of the methods exemplified below.
- Server device 1510A acquires and holds the estimation model generated in ST1104, inputs target behavior data into this estimation model, and acquires target menopausal state data output from this estimation model.
- The server device 1510A transmits the target behavior data via the communication network 2 to an external device (any other server device 1510 or any terminal device 1520) that acquires and holds the estimation model generated in ST1104. to output target menopausal state data to the estimation model that has input the target behavior data as an explanatory variable. After that, the server device 1510A receives the target menopausal state data output by this estimation model from the external device via the communication network 2 .

Acquisition of such target menopausal state data is performed instead of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or managed by the company that operates the estimation service or the like. Along with server device 1510A (and/or terminal device 1520), it may be performed by other server device 1510 (and/or terminal device 1520) managed by other companies.

(5) Step 1108
Next, in ST1108, an information processing device, for example, a server device 1510A (and/or a terminal device 1520) managed by a company that operates an estimation service or the like uses the target menopausal state data acquired in ST1106, At least one target product/one target service corresponding to the target menopausal status data can be determined. Such target products/target services may be appropriate for the current (presumed) menopausal symptom status of the target user identified by the target menopausal status data.

Specifically, in one example, the server device 1510A can first prepare and store a search table that associates and stores target menopausal state data with a plurality of products and/or a plurality of services. Next, server device 1510A inputs the target menopausal condition data acquired in ST1106 into the search table, thereby providing data identifying at least one target product and/or at least one target service from the search table ("proposal data”) can be obtained.

FIG. 28 conceptually shows another example of a search table used in communication system 1500 shown in FIG. The search table exemplified in FIG. 28 is associated with each data (each of the target general condition data, the target first condition data, and the target second condition data) that can be included in the target menopausal condition data, and at least one data to be proposed. stores one target product and/or at least one target service.

In this example, for simplification of explanation, at least one level to be proposed is associated with each of three mutually different levels indicated by each of the target general state data, the target first state data, and the target second state data. One target product and/or at least one target service may be assigned. Taking the target general condition data as an example, the target general condition data indicates class V1 (a class in which the target user's general condition of menopausal symptoms is good), class V2 (a class in which the target user's general condition of menopausal symptoms is standard). At least one target product and/or at least one target service to be proposed are assigned in association with each of the class V3 (the class in which the target user's overall state of menopausal symptoms is poor). obtain.

Classes W1-W3 may also indicate, in this order, the state of the target user's first symptom: "Good", "Normal", and "Poor". Classes X1-X3 may also indicate, in this order, the status of the target user's second symptom "good", "normal", and "bad".

The products exemplified in FIG. 28 (products with reference symbols consisting of “P” and numbers such as P10 and P20) are all arbitrary products suitable for improving the state of menopausal symptoms of the user. , and the services exemplified in FIG. 28 (services with reference signs consisting of “S” and numbers such as S10 and S20) are all suitable for improving the state of menopausal symptoms of the user Indicates any service. These commodities can include, but are not limited to, food, household items, medicines, exercise equipment, electrical appliances, and the like. These services may also include, but are not limited to, hospitals, clinics, gyms, and the like.

For example, when the target comprehensive condition data in the target menopausal condition data indicates class V3, server device 1510A can determine product P40 and service S10 as the target product and target service, respectively. Further, for example, when the target menopause state data includes target second state data (indicating class X2) and target first state data (indicating class W3), server device 1510A sets product P80 as a target product, Services S10, S30, and S40 can be determined as target services.

For each data included in the target menopausal state data, the total number of classes used, the total number of products and/or services assigned to each class, and the like can be determined arbitrarily.

In another example, server device 1510A uses another inference model generated by performing supervised learning (machine learning) to determine at least one target product and/or at least one target service. is also possible. The other estimation model uses data indicating each class of each data included in the target menopausal state data as an explanatory variable, and data indicating at least one target product and/or at least one target service as an objective variable, It can be generated (updated) by performing supervised learning.

Machine learning for training the other estimation model is limited to, for example, any known machine learning, such as gradient boosting trees (e.g., LightGBM), neural networks, or linear regression, as described above. can be included without

By inputting the target menopausal state data into the separate estimation model, the server device 1510A obtains data identifying at least one target product and/or at least one target service ("suggestion data") from the separate estimation model. ) can be output.

Any one of the methods exemplified below can be used as a method of using the other estimation model.
The server device 1510A acquires and holds the generated another estimation model, inputs target menopausal state data into the another estimation model, and acquires proposal data output from the another estimation model. .
The server device 1510A transmits the target menopause via the communication network 2 to an external device (any other server device 1510 or any terminal device 1520) that acquires and holds the generated another estimation model. By transmitting the state data, the other estimation model to which the target menopausal state data is input as an explanatory variable is caused to output proposal data. After that, the server device 1510A receives the proposal data output by the other estimation model from the external device via the communication network 2 .

Acquisition of such proposal data can be performed in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by the server device managed by the company that operates the estimation service or the like. 1510A (and/or terminal 1520) may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

(6) Step 1110
Referring to FIG. 26 again, next, the server device 1510A (and/or the terminal device 1520), which is an information processing device, managed by a company that operates an estimation service, etc., converts the proposal data acquired in ST1108. , can be provided (transmitted) via the communication network 2 to at least one terminal device 1520 (which may be another server device 1510) managed by a corporate user and/or an individual user. "Individual users" as used herein may include "target users."

Proposal data can be provided by corporate users and/or individual users by any means including web pages, e-mails, chats, electronic files, recording media (USB memory, CD-ROM, etc.), and/or paper media. may be provided to at least one terminal device 1520 (which may be another server device 1510) managed by .

Such proposal data is provided in place of the server device 1510A (and/or the terminal device 1520) managed by the company that operates the estimation service or the like, or by the server device managed by the company that operates the estimation service or the like. 1510A (and/or terminal 1520) may be performed by other server devices 1510 (and/or terminal 1520) managed by other companies.

6.7 Modifications In the various examples described above, cases were described in which the sample overall state data, the sample first state data, and the sample second state data were used as objective variables to generate and use an estimation model. In another example, at least one (up to two) of the sample overall state data, the sample first state data, and the sample second state data are used as explanatory variables instead of the objective variable to generate and use the estimation model. It is also possible to

Further, in the various examples described above, the executing entity of the various steps (eg, ST1100 to ST1110) illustrated in FIG. 26 is mainly the server device 1510A (and/or The case of the terminal device 1520) has been described. However, the various steps illustrated in FIG. 26 can also be shared and executed by arbitrary multiple server devices 1510 and/or multiple terminal devices 1520 . Here, any of the plurality of server devices 1510 and/or the plurality of terminal devices 1520 may include, for example, a server device 1510A (and/or the terminal device 1520) managed by a company that operates an estimation service or the like. In another example, server device 1510A (and/or terminal device 1520) managed by a company that operates an estimation service or the like is not included.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Targeted menopausal condition data that identifies the user's current (predicted) menopausal symptom condition can be obtained. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target menopausal state data acquired in this manner is selected by the target user. any user, including the target user, to improve menopausal symptoms.

Thereby, a computer program, an information processing device, which is used for estimating the state of menopausal symptoms of a target user based on the history of at least one action performed by the target user, and which has at least partially improved performance. and methods.

6.8 Estimation Models for Predicting Various Health States It can also be applied to predict the state of general health.
・Exercise status ・Immune status ・Hydration status ・Nutrition status

(1) State of exercise An estimation model for predicting the state of exercise is used for predicting the state related to lifestyle-related diseases. In order to generate such an estimation model, the sample behavior data, which is an explanatory variable, can be at least one of the multiple data exemplified below.
・Data identifying products (or categories to which those products belong) purchased by sample users over a certain period of time (e.g., one year) ・Services used by sample users (or to which such services belong) during a certain period of time (e.g., one year) category) ・Data for identifying whether the sample user is obese ・Data for identifying the amount of change in the weight of the sample user over a certain period of time (for example, from the age of 30 to the present) ・Data identifying the amount of exercise (e.g. steps taken) of the sample user

Also, the data identifying the symptom, which is the objective variable, can be at least one of the plurality of data (which may be measured values) exemplified below.
Data identifying the weight of the sample user Data identifying the BMI (Body Mass Index) of the sample user Data identifying the number of steps taken by the sample user Data identifying smoking of the sample user Diagnosis of metabolic syndrome of the sample user Data identifying the reference value

Note that the metabolic syndrome diagnostic reference value includes the visceral fat accumulation and/or the waist circumference of the sample user, and at least two of the values exemplified below.
Threshold values established for the diagnosis of hyperglyceridemia and/or low HDL cholesterol for the sample user Systolic and/or diastolic blood pressure for the sample user For the diagnosis of fasting hyperglycemia for the sample user threshold set in

The information processing device can generate an estimation model that predicts the state of motion by executing machine learning using such explanatory variables and objective variables. The operations performed by the communication system 1500 in relation to the method of generating and using such an estimation model are described in the above sections "6.6" and "6. 7” can be substantially the same as the operation described in Section 7. In the explanations described in the above "6.6" and "6.7" (especially with reference to FIGS. 26 and 28), the explanatory variables used in the estimation model for predicting the state of menopausal symptoms and Communication related to the method of generating and using an estimation model for predicting the state of motion by replacing the objective variable with the explanatory variable and objective variable described in Section 6.8(1), respectively The operations performed by system 1500 will be understood by those skilled in the art.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Target state data can be obtained that identifies the user's current (predicted) state of motion. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target state data acquired in this manner is selected by the target user. Suggestions can be made to any user, including the target user, to improve their state of exercise.

Thereby, a computer program, an information processing apparatus, and a computer program, having at least partially improved performance, used for estimating the state of motion of a target user based on a history of at least one action performed by the target user; can provide a method.

(2) Immune status Predictive models for predicting immune status are used to predict the severity of symptoms in people whose body is immune-compromised. In order to generate such an estimation model, the sample behavior data, which is an explanatory variable, can be at least one of the multiple data exemplified below.
・Data identifying products (or categories to which those products belong) purchased by sample users over a certain period of time (e.g., one year) ・Services used by sample users (or to which such services belong) during a certain period of time (e.g., one year) data that identifies the category)

Also, the data identifying the symptom, which is the objective variable, can be at least one of the plurality of data (which may be measured values) exemplified below.
・Data identifying sample user's susceptibility to catching a cold ・Data identifying sample user's frequency of catching a cold ・Data identifying sample user's SIgA concentration ・Data identifying sample user's allergic symptoms ・Sample user's oral cavity Data identifying environmental and/or immune status ・Data identifying sleep status of sample users ・Data identifying exercise status of sample users ・Data identifying stress status of sample users

Data for identifying such symptoms may be subjective symptoms of sample users acquired from questionnaires or the like, or may be data acquired from reception histories of sample users at medical institutions. Data identifying such symptoms may identify any of a plurality (preferably three or more) categories.

The information processing device can generate an estimation model that predicts the state of immunity by executing machine learning using such explanatory variables and objective variables. The operations performed by the communication system 1500 in relation to the method of generating and using such an estimation model are described in the above sections "6.6" and "6. 7” can be substantially the same as the operation described in Section 7. In the explanations described in the above "6.6" and "6.7" (especially with reference to FIGS. 26 and 28), the explanatory variables used in the estimation model for predicting the state of menopausal symptoms and By replacing the objective variable with the explanatory variable and objective variable described in Section 6.8(2), respectively, communication related to the method of generating and using an estimation model for predicting the state of immunity, etc. The operations performed by system 1500 will be understood by those skilled in the art.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Subject status data can be obtained that identifies the user's current (predicted) immune status. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target state data acquired in this manner is selected by the target user. It can be offered to any user, including the target user, to improve their immune status.

Thereby, a computer program, an information processing apparatus, and a computer program having at least partially improved performance for estimating the state of immunity of a target user based on a history of at least one action performed by the target user; can provide a method.

(3) Hydration status An estimation model for predicting hydration status is used to predict the severity of symptoms in people with weakened bodies. In order to generate such an estimation model, the sample behavior data, which is an explanatory variable, can be at least one of the multiple data exemplified below.
・Data identifying products (or categories to which those products belong) purchased by sample users over a certain period of time (e.g., one year) ・Services used by sample users (or to which such services belong) during a certain period of time (e.g., one year) data that identifies the category)

Also, the data identifying the symptom, which is the objective variable, can be at least one of the plurality of data (which may be measured values) exemplified below.
Data identifying the sample user's serum Na value Data identifying the sample user's urine osmotic pressure Data identifying the sample user's urine specific gravity Data identifying the sample user's urine color BUN/creatinine ratio of the sample user・Data identifying the sample user’s dehydration rating scale ・Data identifying the sample user’s subjective symptoms associated with dehydration

These data may be generated from laboratory test values obtained from physical examinations or complete physicals received by the sample users, or from the sample users' responses to health questionnaires. Note that the data for identifying subjective symptoms associated with dehydration of the sample users can be generated from the contents of the sample users' answers to the questionnaire. The degree of such subjective symptoms can be grasped from serum Na value, urine osmotic pressure, urine specific gravity, urine color, BUN/creatinine ratio, water intake and/or physical activity.

The information processing device can generate an estimation model that predicts the state of hydration by executing machine learning using such explanatory variables and objective variables. The operations performed by the communication system 1500 in relation to the method of generating and using such an estimation model are described in the above sections "6.6" and "6. 7” can be substantially the same as the operation described in Section 7. In the explanations described in the above "6.6" and "6.7" (especially with reference to FIGS. 26 and 28), the explanatory variables used in the estimation model for predicting the state of menopausal symptoms and By replacing the objective variable with the explanatory variable and objective variable described in Section 6.8 (3), respectively, in relation to the generation and use of an estimation model for predicting the state of hydration The operations performed by communication system 1500 will be understood by those skilled in the art.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Target status data can be obtained that identifies the user's current (predicted) hydration status. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target state data acquired in this manner is selected by the target user. Suggestions can be made to any user, including the target user, to improve their hydration status.

Thereby, a computer program, an information processing device, which is used to estimate the hydration status of a target user based on a history of at least one action performed by the target user, and which has at least partially improved performance. and methods.

(4) Nutritional status Estimation models for predicting nutritional status are used to predict the severity of symptoms in nutritionally deficient individuals. In order to generate such an estimation model, the sample behavior data, which is an explanatory variable, can be at least one of the multiple data exemplified below.
・Data identifying products (or categories to which those products belong) purchased by sample users over a certain period of time (e.g., one year) ・Services used by sample users (or to which such services belong) during a certain period of time (e.g., one year) data that identifies the category)

Also, the data identifying the symptom, which is the objective variable, can be at least one of the plurality of data (which may be measured values) exemplified below.
・Data identifying the presence or absence of subjective symptoms associated with malnutrition of the sample user ・Data identifying the diversity score (DVS) of food intake of the sample user ・Simplified Nutritional Appetite Questionnaire (SNAQ) of the sample user Data to identify ・Data to identify the BMI (Body Mass Index) of the sample user

Data for identifying such symptoms may be subjective symptoms of sample users acquired from questionnaires or the like, or may be data acquired from reception histories of sample users at medical institutions. Data identifying such symptoms may identify any of a plurality (preferably three or more) categories.

The information processing device can generate an estimation model for predicting the state of nutrition by executing machine learning using such explanatory variables and objective variables. The operations performed by the communication system 1500 in relation to the method of generating and using such an estimation model are described in the above sections "6.6" and "6. 7” can be substantially the same as the operation described in Section 7. In the explanations described in the above "6.6" and "6.7" (especially with reference to FIGS. 26 and 28), the explanatory variables used in the estimation model for predicting the state of menopausal symptoms and Communication related to methods of generating and using estimation models for predicting nutritional status by replacing objective variables with explanatory variables and objective variables described in Section 6.8(4), respectively The operations performed by system 1500 will be understood by those skilled in the art.

As described above, according to the technology disclosed in the present application, by inputting the target action data that identifies the history of at least one action performed by the target user into the estimation model, the target Target status data can be obtained that identifies the user's current (predicted) nutritional status. Furthermore, among a plurality of target products and/or a plurality of target services prepared in advance, at least one target product and/or at least one target service suitable for the target state data acquired in this manner is selected by the target user. Suggestions can be made to any user, including the target user, to improve their nutritional status.

Thereby, a computer program, an information processing device, and a computer program having at least partially improved performance for estimating the nutritional status of a target user based on a history of at least one action performed by the target user; can provide a method.

6.10 Various Aspects A computer program according to a first aspect is a computer program that "obtains target action data identifying a history of actions performed by a target user, and performs supervised learning, by being executed by at least one processor. Target sleep state data that identifies the sleep state of the target user by inputting the target behavior data into an estimation model generated by executing At least of property data, target rhythm data identifying the target user's sleep rhythm, target time data identifying the target user's sleep time, and target category data identifying the sleep category to which the target user belongs operable to cause target sleep state data, including one, to be output from the estimation model.

A computer program according to a second aspect is the computer program according to the first aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; obtaining the estimated model generated by inputting sets of teacher data into a learning model, including sample sleep state data identifying sleep states of one sample user corresponding to the sets; the at least one processor may be 'operated as such'.

A computer program according to a third aspect is the computer program according to the first aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; Communicating a plurality of sets of teacher data, including sample sleep state data identifying sleep states of one sample user corresponding to the sets, to the inference model generated by inputting into the learning model. connected via a line;

A computer program according to a fourth aspect is the computer program according to the first aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; sample sleep state data identifying the sleep state of one sample user corresponding to the set; may cause said at least one processor to generate said estimation model.

A computer program according to a fifth aspect is a computer program according to the fourth aspect, wherein "the plurality of sets of acquired teacher data includes a first plurality of sets of teacher data, and is included in the first plurality of sets of teacher data Each set of teacher data includes, as the sample sleep state data, at least sample property data identifying sleep properties of one sample user corresponding to the set, and sleep of one sample user corresponding to the set. and sample rhythm data that identifies the rhythm of, and for each set of teacher data included in the first plurality of sets of teacher data that has been acquired, one sample user's sleep characteristics corresponding to the set A sample sleep identifying a sleep category to which one sample user corresponding to the set belongs based on the sample disposition data to identify and sample rhythm data identifying sleep rhythms of the one sample user corresponding to the set. Preprocessing is performed to generate category data, and each set of teacher data is composed of the sample behavior data of one sample user corresponding to the set and the sample behavior data of one sample user corresponding to the set. Input the first plurality of sets of teacher data after preprocessing, including sleep category data, into the learning model to generate the estimation model, and from the estimation model, as the target sleep state data, The at least one processor may be caused to output target sleep category data identifying a sleep category to which the target user belongs.

A computer program according to a sixth aspect is a computer program according to any one of the second to fourth aspects, wherein "the plurality of sets of teacher data includes a second plurality of sets of teacher data, and the second plurality of sets of teacher data Each set of teacher data included in the teacher data includes, as the sample sleep state data, at least sample property data identifying the sleep property of one sample user corresponding to the set and one person corresponding to the set sample sleep category data identifying the sleep category to which one sample user corresponding to the set belongs, determined based on the sample rhythm data identifying the sleep rhythms of the sample users of the inference model; and cause the at least one processor to output, as the target sleep state data, target sleep category data identifying a sleep category to which the target user belongs.

In the computer program according to the seventh aspect, in the fifth aspect or the sixth aspect, "the sample property data that identifies the sleep property of one sample user corresponding to the set is Athens insomnia Scale, Pittsburgh sleep questionnaire, three-dimensional sleep scale, or score calculated based on the content of the sample user's answers to the insomnia severity questionnaire, showing the sleep of one sample user corresponding to the set The sample rhythm data that identifies the rhythm of indicates social jet lag, which is the absolute value of the difference between the median time of the sample user's sleep time on weekdays and the median time of sleep on holidays, and corresponds to the set The sleep category to which one sample user belongs has at least one threshold value arranged in correspondence with the score on one of the horizontal axis and the vertical axis, and the other of the horizontal axis and the vertical axis and at least one threshold associated with said social jet lag on the axis of .

A computer program according to an eighth aspect is a computer program according to any one of the second aspect to the fourth aspect, wherein "the plurality of sets of teacher data includes a third plurality of sets of teacher data, and the third plurality of sets of teacher data Each set of teacher data included in the teacher data includes, as the sample sleep state data, at least sample property data identifying the sleep property of one sample user corresponding to the set, and one corresponding to the set sample rhythm data that identifies sleep rhythms of a human sample user, and from the estimation model, as the target sleep state data, target property data that identifies the sleep properties of the target user; operable to output subject rhythm data identifying sleep rhythms.

A computer program according to a ninth aspect is, in any one of the second aspect to the fourth aspect, "the plurality of sets of teacher data includes a fourth plurality of sets of teacher data, and the fourth plurality of sets of teacher data Each set of teacher data included in the teacher data includes, as the sample sleep state data, at least sample property data that identifies the sleep property of one sample user corresponding to the set, and from the estimation model, The at least one processor may be operable to output, as the targeted sleep state data, targeted disposition data identifying a sleep disposition of the targeted user.

A computer program according to a tenth aspect is a computer program according to any one of the second aspect to the fourth aspect, wherein "the plurality of sets of teacher data includes a fifth plurality of sets of teacher data, and the fifth plurality of sets of teacher data Each set of teacher data included in the teacher data includes, as the sample sleep state data, at least sample rhythm data that identifies the sleep rhythm of one sample user corresponding to the set, and from the estimation model, The at least one processor may function to output, as the target sleep state data, target rhythm data identifying sleep rhythms of the target user.

A computer program according to an eleventh aspect is a computer program according to any one of the second aspect to the fourth aspect, wherein "the plurality of sets of teacher data includes a sixth plurality of sets of teacher data, and the sixth plurality of sets of teacher data Each set of teacher data included in the teacher data includes at least sample time data identifying the sleep time of one sample user corresponding to the set as the sample sleep state data, and from the estimation model, the operable to cause the at least one processor to output, as targeted sleep state data, targeted time data identifying sleep times of the targeted user.

A computer program according to a twelfth aspect is a computer program according to any one of the first to eleventh aspects, wherein "out of a plurality of products and/or a plurality of services, at least one target corresponding to the target sleep state data The at least one processor may be operable to determine goods and/or at least one target service.

A computer program according to a thirteenth aspect, wherein in the twelfth aspect, "by inputting the target sleep state data into another estimation model generated by performing supervised learning, from the another estimation model , cause said at least one processor to output suggestion data identifying said at least one target product and/or said at least one target service.

A computer program according to a fourteenth aspect is the computer program according to the twelfth aspect, wherein "inputting the target sleep state data into a search table that associates the target sleep state data with the plurality of products and/or the plurality of services may cause the at least one processor to obtain suggestion data identifying the at least one target product and/or the at least one target service from the lookup table.

In the computer program according to the fifteenth aspect, in any one of the first to fourteenth aspects, "the at least one processor comprises a central processing unit (CPU), a microprocessor, and/or may include a graphics processing unit (GPU).

An information processing apparatus according to a sixteenth aspect "comprising at least one processor, the at least one processor acquires target action data identifying a history of actions performed by a target user, and performs supervised learning. Target sleep state data for identifying a sleep state of the target user by inputting the target behavior data into an estimation model generated by executing a target property for identifying a sleep property of the target user data, target rhythm data identifying the target user's sleep rhythm, target time data identifying the target user's sleep time, and target category data identifying the sleep category to which the target user belongs. outputting target sleep state data including one from the estimation model;
configured to 'get'.

The information processing device according to the seventeenth aspect may be "a terminal device or a server device" in the sixteenth aspect.

In the information processing apparatus according to the eighteenth aspect, in the seventeenth aspect, "the at least one processor comprises a central processing unit (CPU), a microprocessor, and/or a graphics processing unit (GPU). can be included.

A method according to the nineteenth aspect is described as "a method performed by at least one processor executing computer-readable instructions, wherein the at least one processor, by executing the instructions, causes a target user to: Obtaining target behavior data identifying a history of actions performed; and inputting the target behavior data into an inference model generated by performing supervised learning to determine the sleep state of the target user. Target sleep state data to be identified, which includes target property data for identifying the sleep property of the target user, target rhythm data for identifying the sleep rhythm of the target user, and target time data for identifying the sleep time of the target user. and target category data identifying a sleep category to which the target user belongs.

In the method according to the twentieth aspect, in the nineteenth aspect, "the at least one processor includes a central processing unit (CPU), a microprocessor, and/or a graphics processing unit (GPU)." can see

A method according to the twenty-first aspect is described as "a method performed by at least one processor executing computer-readable instructions, wherein the at least one processor executes the instructions to cause each set of The teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set, and sample sleep state data identifying the sleep state of the one sample user corresponding to the set. and target action data identifying a history of actions performed by the target user by inputting the plurality of sets of teacher data into a learning model and making it learn. to input target sleep state data for identifying the sleep state of the target user, which includes target property data for identifying the sleep property of the target user and target rhythm data for identifying the sleep rhythm of the target user. , target time data identifying sleep duration of the target user, and target category data identifying a sleep category to which the target user belongs. and a generating stage that generates an estimated model.

A method according to a twenty-second aspect is the method according to the twenty-first aspect, wherein "the obtaining step includes obtaining a first plurality of sets of teacher data included in the plurality of sets of teacher data; Each set of teacher data included in the set of teacher data includes, as the sample sleep state data, at least sample property data identifying the sleep property of one sample user corresponding to the set, and sample property data corresponding to the set. and sample rhythm data identifying sleep rhythms of a sample user, and wherein said generating step includes, for each set of teacher data included in said first plurality of sets of acquired teacher data, said Based on sample property data identifying sleep properties of one sample user corresponding to the set and sample rhythm data identifying sleep rhythms of one sample user corresponding to the set, performing pre-processing to generate sample sleep category data that identifies the sleep category to which a sample user belongs; data and the sample sleep category data of one sample user corresponding to the set, inputting the first plurality of sets of teacher data after preprocessing into the learning model to input the estimation model and generating a .

A method according to a twenty-third aspect is the method according to the twenty-first aspect, wherein "the obtaining step includes obtaining a second plurality of sets of teacher data included in the plurality of sets of teacher data; Each set of teacher data included in the set of teacher data includes, as the sample sleep state data, at least sample property data identifying the sleep property of one sample user corresponding to the set and one corresponding to the set. sample sleep category data identifying a sleep category to which one sample user corresponding to the set belongs, determined based on the sample rhythm data identifying sleep rhythms of the sample users; inputting a plurality of sets of training data from said learning model to generate said estimation model.

In the method according to the twenty-fourth aspect, in the twenty-third aspect above, "the sample property data identifying sleep properties of one sample user corresponding to the set comprises: Athens Insomnia Scale, Pittsburgh Sleep Questionnaire; A sample rhythm that identifies the sleep rhythm of one sample user corresponding to the set, showing a score calculated based on the content of the sample user's answers to the three-dimensional sleep scale or the insomnia severity questionnaire. The data show the social jet lag, which is the absolute value of the difference between the median time of the sample user's sleep time on weekdays and the median time of sleep time on holidays, and one sample user corresponding to the set The sleep category to which it belongs includes at least one threshold value arranged in correspondence with the score on one of the horizontal and vertical axes, and the social jet lag on the other of the horizontal and vertical axes. and at least one threshold value arranged in correspondence with a plurality of categories divided by.

A method according to a twenty-fifth aspect is, in any one of the twenty-first to the twenty-fourth aspects, "By inputting target action data identifying a history of actions performed by the target user into the estimation model, an estimation step of outputting target sleep state data identifying states of sleep of the target user from the estimation model.

In the method according to the twenty-sixth aspect, in any one of the twenty-first to the twenty-fifth aspects, "the at least one processor is a central processing unit (CPU), a microprocessor, and/or a graphics may include a processing unit (GPU).

A computer program according to a twenty-seventh aspect is "executed by at least one processor to acquire target action data identifying a history of actions performed by a target user, and generate by performing supervised learning target climacteric state data for identifying the state of the climacteric symptoms of the target user by inputting the target behavior data into the estimated model, the target synthesis for identifying the comprehensive state of the climacteric symptoms of the target user a target menopausal condition including at least one of condition data, target first condition data identifying a condition of a first symptom of the target user, and target second condition data identifying a condition of a second symptom of the target user The at least one processor may be 'operated to cause data to be output from the estimation model'.

A computer program according to a twenty-eighth aspect is the computer program according to the twenty-seventh aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; obtaining the estimation model generated by inputting a plurality of sets of teacher data into a learning model, including sample menopausal state data identifying a menopausal symptom state of one sample user corresponding to the sets; , the at least one processor may be "operated as".

A computer program according to a twenty-ninth aspect is the computer program according to the twenty-seventh aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; sample menopausal status data identifying the menopausal symptom status of one sample user corresponding to the set; connected via a communication line;

A computer program according to a thirtieth aspect is the computer program according to the twenty-seventh aspect, wherein "each set of teacher data includes sample action data identifying a history of actions performed by one sample user corresponding to the set; and sample menopausal state data identifying the state of menopausal symptoms of one sample user corresponding to the set. thereby causing said at least one processor to generate said estimation model.

A computer program according to a thirty-first aspect is the computer program according to the thirtieth aspect, wherein "the plurality of sets of acquired teacher data includes a first plurality of sets of teacher data, and is included in the first plurality of sets of teacher data each set of teacher data includes, as the sample menopausal state data, at least sample overall state data identifying an overall state of menopausal symptoms of one sample user corresponding to the set; from the estimation model; outputting, as the target menopausal state data, target comprehensive state data identifying a comprehensive state of the target user's menopausal symptoms;
the at least one processor may be 'operated as such'.

A computer program according to a thirty-second aspect is the computer program according to the thirty-first aspect, wherein "each set of teacher data included in the first plurality of sets of teacher data corresponds to at least the set as the sample menopausal state data for each set of teacher data included in the first plurality of sets of teacher data, including sample overall condition data identifying a score corresponding to the overall condition of menopausal symptoms of one sample user, corresponding to the set; sample overall condition data identifying a score corresponding to the overall condition of the menopausal symptoms of the one sample user corresponding to the set of categories corresponding to the overall condition of the menopausal symptoms of the one sample user corresponding to the set; performing preprocessing of converting into sample comprehensive state data to be identified, inputting the first plurality of sets of teacher data after the preprocessing into the learning model to generate the estimation model, and generating the estimation model; the at least one processor to output, as the targeted menopausal state data, targeted aggregated condition data identifying a category corresponding to an aggregated condition of the target user's menopausal symptoms from .

In the computer program according to the thirty-third aspect, in the thirty-second aspect, "the score corresponding to the comprehensive state of menopausal symptoms of one sample user corresponding to the group is a simplified menopausal index (SMI) 、更年期障害評価尺度（Ｍｅｎｏｐａｕｓｅ Ｒａｔｉｎｇ Ｓｃａｌｅ）、クッパーマン更年期指数（Ｋｕｐｐｅｒｍａｎ ｉｎｄｅｘ）、更年期障害評価（Ｇｒｅｅｎ Ｃｌｉｍａｃｔｅｒｉｃ Ｓｃａｌｅ）、ＰＳＳＴ（Ｐｒｅｍｅｎｓｔｒｕａｌ Ｓｙｍｐｔｏｍｓ Ｓｃｒｅｅｎｉｎｇ Ｔｏｏｌ）、ＷＨＱ（Ｗｏｍｅｎ'ｓ Ｈｅａｌｔｈ Ｑｕｅｓｔｉｏｎｎａｉｒｅ）、ＶＡＳ（Ｖｉｓｕａｌ Ａｎａｌｏｇｕｅ Ｓｃａｌｅ ), HFRDI (Hot Flash Related Daily Interference Scale), HFCS (Hot Flash Composite Score), MENQOL (Menopause-Specific Quality Of Life) or Japanese women's menopausal symptom evaluation table. score.

A computer program according to a thirty-fourth aspect is the computer program according to the thirtieth aspect, wherein "the plurality of sets of acquired teacher data includes a second plurality of sets of teacher data, and is included in the second plurality of sets of teacher data Each set of teacher data includes, as the sample menopausal state data, at least sample first state data identifying a first symptom state of one sample user corresponding to the set, and from the estimation model, the The at least one processor may be caused to output, as targeted menopausal condition data, targeted first condition data identifying a first symptom condition of the targeted user.

A computer program according to a thirty-fifth aspect is the computer program according to the thirty-fourth aspect, wherein "each set of teacher data included in the second plurality of sets of teacher data corresponds to at least the set as the sample menopausal state data For each set of teacher data included in said second plurality of sets of teacher data, including sample first condition data identifying a score corresponding to a first symptom condition of one sample user, corresponding to said set Sample first condition data identifying a score corresponding to a first symptom condition of a sample user; performing preprocessing of converting into one-state data, inputting the second plurality of sets of teacher data after the preprocessing into the learning model to generate the estimated model, and generating the estimated model from the estimated model; operable to cause the at least one processor to output, as targeted menopausal condition data, targeted first condition data identifying a category corresponding to a first symptom condition of the targeted user.

A computer program according to a thirty-sixth aspect is a computer program according to the thirtieth aspect, wherein "the plurality of sets of acquired teacher data includes a third plurality of sets of teacher data, and is included in the third plurality of sets of teacher data Each set of teacher data includes, as the sample menopausal state data, at least sample second state data that identifies a second symptom state of one sample user corresponding to the set, and from the estimation model, the The at least one processor may be caused to output, as targeted menopausal condition data, targeted first condition data identifying a second symptom condition of the targeted user.

A computer program according to a thirty-seventh aspect is the computer program according to the thirty-sixth aspect, wherein "each set of teacher data included in the third plurality of sets of teacher data corresponds to at least the set as the sample menopausal state data For each set of teacher data included in the third plurality of sets of teacher data, including sample second condition data identifying a score corresponding to a condition of the second symptom of one sample user, corresponding to the set Sample second condition data identifying a score corresponding to a second symptom condition of a sample user; performing preprocessing of converting into two-state data, inputting the third plurality of sets of teacher data after the preprocessing into the learning model to generate the estimated model, and generating the estimated model from the estimated model; operable to cause the at least one processor to output, as targeted menopausal condition data, targeted second condition data identifying a category corresponding to a second symptom condition of the targeted user.

A computer program according to a thirty-eighth aspect, according to any one of the twenty-seventh to thirty-seventh aspects, wherein "out of a plurality of products and/or a plurality of services, at least one target corresponding to the target menopausal state data The at least one processor may be operable to determine goods and/or at least one target service.

A computer program according to a thirty-ninth aspect, wherein in the thirty-eighth aspect, "by inputting the target menopausal state data into another estimation model generated by performing supervised learning, from the another estimation model, , cause said at least one processor to output suggestion data identifying said at least one target product and/or said at least one target service.

A computer program according to a 40th aspect is the computer program according to the 38th aspect, wherein "inputting the target menopausal state data into a search table that associates the target menopausal state data with the plurality of products and/or the plurality of services may cause the at least one processor to obtain suggestion data identifying the at least one target product and/or the at least one target service from the lookup table.

In the computer program according to the 41st aspect, in any one of the 27th to 40th aspects, "the at least one processor comprises a central processing unit (CPU), a microprocessor, and/or It may include a graphics processing unit (GPU).

An information processing apparatus according to a forty-second aspect "comprising at least one processor, the at least one processor acquiring target action data identifying a history of actions performed by a target user, and performing supervised learning. target menopausal condition data identifying a state of the target user's menopausal symptoms by inputting the target behavior data into an estimation model generated by executing the target climacteric condition data, wherein the overall condition of the target user's climacteric symptoms target comprehensive condition data for identifying the target user, target first condition data for identifying the condition of the first symptom of the target user, and target second condition data for identifying the condition of the target user's second symptom, at least It may be configured to cause target menopausal status data, including one, to be output from the estimation model.

The information processing device according to the forty-third aspect may be "a terminal device or a server device" in the forty-second aspect.

In the information processing apparatus according to the forty-fourth aspect, in the forty-second aspect or the forty-third aspect, "the at least one processor comprises a central processing unit (CPU), a microprocessor, and/or graphics It may include a processing unit (GPU).

A method according to the forty-fifth aspect is described as "a method performed by at least one processor executing computer-readable instructions, wherein said at least one processor, by executing said instructions, causes a target user to: Menopausal symptom status of the target user by obtaining target behavior data identifying a history of behaviors performed and inputting the target behavior data into an inference model generated by performing supervised learning. which is target comprehensive condition data for identifying the overall condition of the target user's menopausal symptoms, target first condition data for identifying the condition of the target user's first symptom, and and causing target menopausal status data including at least one of: target second condition data identifying a second symptom status of the target user to be output from the estimation model.

In the method according to the forty-sixth aspect, in the forty-fifth aspect, "the at least one processor includes a central processing unit (CPU), a microprocessor, and/or a graphics processing unit (GPU)." can see

A method according to the forty-seventh aspect is described as "a method performed by at least one processor executing computer-readable instructions, the at least one processor executing the instructions to cause each set of The teacher data includes sample behavior data identifying a history of behaviors performed by one sample user corresponding to the set, and sample menopausal state identifying a menopausal symptom state of the one sample user corresponding to the set. a target action that identifies a history of actions performed by the target user by inputting and learning the plurality of sets of teacher data into a learning model, comprising: Target climacteric condition data for identifying a condition of the climacteric symptoms of the target user by inputting data, the target comprehensive condition data for identifying an overall condition of the climacteric symptoms of the target user; to output target menopausal state data including at least one of target first state data identifying a symptom state and target second state data identifying a second symptom state of the target user and a generating stage for generating the constructed estimation model.

A method according to a forty-eighth aspect is the method according to the forty-seventh aspect, wherein "the obtaining step includes obtaining a first plurality of sets of teacher data included in the plurality of sets of teacher data, Each set of teacher data included in the set of teacher data includes, as the sample menopausal state data, at least a sample synthesis that identifies a score corresponding to the comprehensive state of menopausal symptoms of one sample user corresponding to the set. condition data, and wherein the generating step generates a comprehensive menopausal symptom of one sample user corresponding to each set of teacher data included in the first plurality of sets of teacher data acquired; transforming the sample aggregate condition data identifying scores corresponding to the conditions into sample aggregate condition data identifying categories corresponding to the aggregate condition of the menopausal symptoms of one sample user corresponding to the set; performing a process, wherein each set of teacher data includes the sample behavior data of one sample user corresponding to the set and the sample general condition data of the one sample user corresponding to the set; and inputting said first plurality of sets of teacher data after said preprocessing into said learning model to generate said estimation model.

A method according to a forty-ninth aspect is the method according to the forty-seventh aspect, wherein "the obtaining step includes obtaining a first plurality of sets of teacher data included in the plurality of sets of teacher data; Each set of teacher data included in the set of teacher data includes, as the sample menopausal state data, at least a sample first state identifying a score corresponding to the state of the first symptom of one sample user corresponding to the set. data, and wherein the generating step includes, for each set of teacher data included in the acquired second plurality of sets of teacher data, the state of the first symptom of one sample user corresponding to the set; Performing pre-processing to convert sample first condition data identifying corresponding scores into sample first condition data identifying categories corresponding to first symptom states of one sample user corresponding to the set. each set of teacher data includes the sample behavior data of one sample user corresponding to the set and the sample first state data of the one sample user corresponding to the set; inputting said second plurality of sets of teacher data after said preprocessing into said learning model to generate said estimation model.

The method according to the fiftieth aspect is the method according to the forty-seventh aspect, wherein "the obtaining step includes obtaining a third plurality of sets of teacher data included in the plurality of sets of teacher data, Each set of teacher data included in the set of teacher data includes, as the sample menopausal state data, at least a sample second state identifying a score corresponding to the second symptom state of one sample user corresponding to the set. data, and wherein the generating step includes, for each set of teacher data included in the acquired third plurality of sets of teacher data, the state of the second symptom of one sample user corresponding to the set Performing pre-processing to convert sample second condition data identifying corresponding scores into sample second condition data identifying categories corresponding to second symptom states of one sample user corresponding to the set. each set of teacher data includes the sample behavior data of one sample user corresponding to the set and the sample second state data of the one sample user corresponding to the set; inputting said third plurality of sets of teacher data after said preprocessing into said learning model to generate said estimation model.

In the method according to the fifty-first aspect, in any of the forty-seventh to fiftieth aspects, "the at least one processor is a central processing unit (CPU), a microprocessor, and/or a graphics processing unit (GPU).

As will be readily appreciated by one of ordinary skill in the art having the benefit of this disclosure, the various examples described above can be used in suitable combinations in various patterns with each other, unless inconsistent.

In view of the many possible embodiments in which the principles of the invention disclosed herein may be applied, the various illustrated embodiments are merely various preferred examples and techniques for the claimed invention. It should be understood that the scope should not be considered limited to these preferred various examples. Indeed, the scope of the claimed invention is defined by the appended claims. Therefore, all matters falling within the technical scope of the invention described in the claims are claimed as inventions of the present inventors.

1: Health prediction system 9: Learning system 100: Integrated health prediction program 110: Health prediction model 120: Input unit 125: Product category determination unit 130: Output unit 135: Proposal unit 140: Output unit 150: Purchase data 160: Health prediction Value 800: Information processing device 810: Control unit 820: Input interface (I/F) unit 830: Storage device 831: Proposal table 840: Output I/F unit 900: Model learning program 910: Input unit 912: Product category determination Unit 914: Learning unit 916: Output unit 950: Teacher data 952: Purchase data 954: Health data 1400: Information processing device 1410: Control unit 1420: Communication I/F unit 1430: Storage device D: Display

Claims (24)

A process of acquiring first purchase data representing the purchase history of the target customer and including information on the product category to which the purchased product belongs;
Prediction processing for predicting the degree of multiple types of health symptoms using a health prediction model for multiple types of health symptoms of types 1 to N (N is an integer of 2 or more),
The health prediction model for the kth (1 ≤ k ≤ N) health symptoms indicates the purchase history of a plurality of subjects, and includes information indicating the product category to which the purchased product belongs. Of the second purchase data, the kth select a product category to be used for predicting the health symptoms of the above, use the purchase data of the selected product category from the second purchase data as an explanatory variable, and use the k-th type of health symptoms collected from each of the plurality of subjects as an explanatory variable It was constructed by machine learning with a numerical value indicating the degree as the objective variable,
The health prediction model for the kth health symptom is a numerical value indicating the degree of the kth health symptom from the purchase data of the product category selected for the health prediction model in the first purchase data. a prediction process that outputs
and providing the target customer with health information corresponding to the numerical values indicating the severity of the plurality of types of health symptoms output by the prediction process. 2. The program according to claim 1, wherein said machine learning is machine learning by logistic regression. 3. The program according to claim 2, wherein said logistic regression uses L1 regularization method and/or L2 regularization method. Health symptoms of the kth type (1 ≤ k ≤ N) are classified into multiple classes according to the degree of health symptoms,
Each of the plurality of classes is associated in advance with a numerical range according to the degree of health symptoms,
In the prediction process, based on the numerical range to which the numerical value indicating the degree of health symptoms of the target customer belongs, predicting the degree of health symptoms of the target customer by determining a class corresponding to the numerical value;
4. The program according to any one of claims 1 to 3. 5. The program according to claim 4, wherein said plurality of classes is three or more. By referring to a table that associates each of the plurality of classes with services and/or products that match the health symptoms of each class, together with the health information, services and products that match the determined class of health symptoms. / or a proposal process for making product recommendations;
6. The program according to claim 4 or 5, further causing the information processing apparatus to execute: 7. The insurance product related to the health information is proposed in the proposal processing by referring to a table that associates each of the plurality of classes with a service that matches the health condition of each class. program. In the proposal processing, a medical institution that performs preventive care and/or treatment related to the health information by referring to a table that associates each of the plurality of classes with a service that matches the health condition of each class. 7. The program according to claim 6, wherein the program introduces to the target customer. The first purchase data is the purchase history of food and/or daily necessities purchased during a certain period, and the degree of k-th (1≦k≦N) health symptoms is the simplified menopausal index (SMI), menopausal disorder evaluation.尺度（ｔｈｅ Ｍｅｎｏｐａｕｓｅ Ｒａｔｉｎｇ Ｓｃａｌｅ）、クッパーマン更年期指数（Ｔｈｅ Ｋｕｐｐｅｒｍａｎ ｉｎｄｅｘ）、更年期障害評価尺度（Ｇｒｅｅｎ Ｃｌｉｍａｃｔｅｒｉｃ Ｓｃａｌｅ）、ＰＳＳＴ（Ｔｈｅ ｐｒｅｍｅｎｓｔｒｕａｌ ｓｙｍｐｔｏｍｓｓｃｒｅｅｎｉｎｇ ｔｏｏｌ）、ＭＲＳ（Ｍｅｎｏｐａｕｓｅ ｒａｔｉｎｇ Ｓｃａｌｅ）、ＷＨＱ（Ｔｈｅ Ｗｏｍｅｎ'ｓ Ｈｅａｌｔｈ Ｑｕｅｓｔｉｏｎｎａｉｒｅ ), VAS (Visual analogue scale), HFRDI (Hot Flash Related Daily Interference Scale), HFCS (Hot Flash Composite Score), and MENQOL (Menopause-Specific Quality of life evaluation of any symptom of Japanese women) represented by one or more
A program according to any one of claims 1 to 8. The target customer's health symptoms include one or more selected from menopausal symptoms, premenstrual syndrome (PMS, PMDD) symptoms,
The health prediction models for the plurality of types of health symptoms include one or more of a menopausal prediction model, a premenstrual syndrome (PMS, PMDD) prediction model,
10. The program according to any one of claims 1 to 9. The degree of the kth type (1 ≤ k ≤ N) health symptoms is represented by any one or more of worsening sleep condition, disturbed sleep rhythm, and inappropriate sleep time.
A program according to any one of claims 1 to 8. 12. The program according to any one of claims 1 to 11, wherein the health prediction models for the plurality of types of health symptoms include sleep state prediction models. 9. Any one of claims 1 to 8, wherein the degree of the kth type (1≤k≤N) health condition is represented by any one or more of lack of exercise, obesity, and change in body weight over a certain period of time. Program described in paragraph. 4. The program according to any one of claims 1 to 3, wherein the machine learning is performed including one or more information of smoking status and sedentary behavior of each of the plurality of subjects as explanatory variables. 15. The program according to any one of claims 1 to 14, wherein the health prediction models for the plurality of types of health symptoms include a lifestyle-related disease prediction model for exercise states. Claims 1 to 4, wherein the degree of the kth type (1 ≤ k ≤ N) health symptoms is represented by any one or more of serum Na, BUN/creatinine ratio, urine osmotic pressure, urine color, and urine specific gravity. Item 9. The program according to any one of Items 8. As an explanatory variable, including one or more information of serum Na, BUN / creatinine ratio, urine osmotic pressure, urine specific gravity, amount of water intake, amount of alcohol consumption, amount of physical activity, color of urine color for each of the plurality of subjects 4. The program according to any one of claims 1 to 3, which performs machine learning. 18. The program according to any one of claims 1 to 17, wherein the health prediction models for the plurality of types of health symptoms include a hydration condition prediction model for hydration conditions. The degree of the kth type (1 ≤ k ≤ N) health symptoms is the susceptibility to colds, frequency of causing colds, SIgA concentration, allergic symptoms, oral environment / immune status, exercise status, stress status, 9. The program according to any one of claims 1 to 8, represented by any one or more of severe sleep symptoms, moderate sleep symptoms, mild sleep symptoms, and cold-like symptoms. 20. The program according to any one of claims 1 to 19, wherein the health prediction models for the plurality of types of health symptoms include an immune state prediction model for immune states. The degree of health symptoms of the kth type (1 ≤ k ≤ N) is the presence or absence of subjective symptoms associated with malnutrition, diversity score of food intake (DVS), Simplified nutritional appetite questionnaire (SNAQ), BMI 9. The program according to any one of claims 1 to 8, represented by any one or more of (Body Mass Index). 22. The program according to any one of claims 1 to 21, wherein the health prediction models for the plurality of types of health symptoms include a nutritional status prediction model for nutritional status. an input unit for obtaining first purchase data representing the purchase history of the target customer and including information on the product category to which the purchased product belongs;
A prediction unit that predicts the degree of the plurality of types of health symptoms using a health prediction model for a plurality of types of health symptoms of types 1 to N (N is an integer of 2 or more),
The health prediction model for the kth (1 ≤ k ≤ N) health symptoms indicates the purchase history of a plurality of subjects, and includes information indicating the product category to which the purchased product belongs. Of the second purchase data, the kth select a product category to be used for predicting the health symptoms of the above, use the purchase data of the selected product category from the second purchase data as an explanatory variable, and use the k-th type of health symptoms collected from each of the plurality of subjects as an explanatory variable It was constructed by machine learning with a numerical value indicating the degree as the objective variable,
The health prediction model for the kth health symptom is a numerical value indicating the degree of the kth health symptom from the purchase data of the product category selected for the health prediction model in the first purchase data. a prediction unit that outputs
and an output unit that provides the target customer with health information corresponding to the numerical values indicating the degree of the plurality of types of health symptoms output by the prediction unit. A process of acquiring first purchase data representing the purchase history of the target customer and including information on the product category to which the purchased product belongs;
A process of predicting the degree of the plurality of types of health symptoms using a health prediction model for a plurality of types of health symptoms of types 1 to N (N is an integer of 2 or more),
The health prediction model for the kth (1 ≤ k ≤ N) health symptoms indicates the purchase history of a plurality of subjects, and includes information indicating the product category to which the purchased product belongs. Of the second purchase data, the kth select a product category to be used for predicting the health symptoms of the above, use the purchase data of the selected product category from the second purchase data as an explanatory variable, and use the k-th type of health symptoms collected from each of the plurality of subjects as an explanatory variable It is constructed by machine learning using a numerical value indicating the degree as an objective variable, and the health prediction model for the k-th type health condition is selected for the health prediction model in the first purchase data. Prediction processing for outputting a numerical value indicating the degree of the kth type health symptom from the purchase data of the product category obtained;
An information processing method in which an information processing device performs a process of providing the target customer with health information corresponding to the numerical values indicating the degree of the plurality of types of health symptoms output by the prediction process.

Images