JP2023131829A - Information processing apparatus, information processing method, and program - Google Patents

Abstract

To provide an information processing apparatus, method, program which obtains a selection probability of a user.SOLUTION: In an information processing system 1000 in which a server device 100 and a plurality of client devices 110 are connected so that they can communicate with each other via a network 150. The server device 100 being the information processing apparatus includes a control unit. The control unit acquires a plurality of pieces of evaluation value information and obtains a selection probability of a user on the basis of the plurality of pieces of obtained evaluation value information and a probability model.SELECTED DRAWING: Figure 1

Description

Application for application of Article 30, Paragraph 2 of the Patent Act filed on October 13, 2021 at https://arxiv. org/abs/2110.06523v1, https://arxiv. org/abs/2110.06523v2, and https://arxiv.in “arXiv” on November 17, 2021. Published at org/abs/2110.06523v3

The present invention relates to an information processing device, an information processing method, and a program.

There is a technology that determines the popularity of a tourist spot in real time and displays tourist information including the popularity (see Patent Document 1).

Japanese Patent Application Publication No. 2012-256303

However, the technique disclosed in Patent Document 1 has a problem in that tourist information that is not popular but that the user prefers is not displayed.

According to one aspect of the present invention, an information processing device is provided. This information processing device has a control section. The control unit acquires a plurality of pieces of evaluation value information. A user's selection probability is determined based on the acquired plurality of evaluation value information and the probability model.

FIG. 1 is a diagram illustrating an example of a system configuration of an information processing system 1000. FIG. 2 is a diagram showing an example of the hardware configuration of the server device 100. FIG. 3 is a diagram showing an example of the hardware configuration of the client device 110. FIG. 4 is a diagram showing the main symbology used in this specification. FIG. 5 is a diagram showing an example of B-UEM. FIG. 6 is a diagram illustrating an example of an algorithm by which the control unit 201 generates a probabilistic model. FIG. 7 is a diagram illustrating an example of ST-UEM. FIG. 8 is a diagram showing an example of S-UEM. FIG. 9 is a diagram showing an example of T-UEM. FIG. 10 is an activity diagram showing an example of information processing in the server device 100. FIG. 11 is a diagram showing an example of a login screen 800. FIG. 12 is a diagram showing an example of the questionnaire screen 900. FIG. 13 is a diagram showing an example of the recommendation screen 1010.

Embodiments of the present invention will be described below with reference to the drawings. Various features shown in the embodiments described below can be combined with each other.

Embodiments of the present invention will be described below with reference to the drawings. Various features shown in the embodiments described below can be combined with each other. In particular, in this specification, the term "unit" may include, for example, a combination of hardware resources implemented by circuits in a broad sense and software information processing that can be specifically implemented by these hardware resources. . In addition, various information is handled in this embodiment, and this information is expressed as a binary bit collection consisting of 0 or 1 by the high/low of the signal value, or by the physical numerical value of the signal value. Communication and computation can be performed on circuits in a broad sense, whether represented by quantum superposition or by quantum superposition.

Further, a circuit in a broad sense is a circuit realized by at least appropriately combining a circuit, a circuit, a processor, a memory, and the like. That is, application specific integrated circuits (ASIC), programmable logic devices (for example, simple programmable logic devices (SPLD), complex programmable logic devices (COMPLEX), etc. Programmable Logic Device (CPLD), and field This includes a field programmable gate array (FPGA) and the like.

Furthermore, the program for realizing the software that appears in the embodiments may be implemented in a form that can be downloaded from a server, the program may be executed on a cloud computer, or it may be a non-volatile or volatile program. It may be stored and distributed on a non-transitory storage medium.

<Embodiment 1>
1. System Configuration FIG. 1 is a diagram showing an example of the system configuration of an information processing system 1000. As shown in FIG. 1, the information processing system 1000 includes a server device 100 and a client device 110 as a system configuration. Server device 100 and client device 110 are communicably connected via network 150. The server device 100 executes the processing of this embodiment described below in response to a request from the client device 110 or the like. The client device 110 is a device operated by a user and, for example, responds to a questionnaire or displays the results of processing by the server device 100.

In FIG. 1, only one client device 110 is shown in the information processing system 1000 to simplify the explanation. However, the information processing system 1000 includes as many client devices 110 as there are users.
Further, in FIG. 1, a smartphone is shown as an example of the client device 110, but the client device 110 is not limited to a smartphone, and may be a PC (Personal Computer) or a tablet computer. Server device 100 is an example of an information processing device.

2. Hardware Configuration (1) Hardware Configuration of Server Device 100 FIG. 2 is a diagram showing an example of the hardware configuration of the server device 100. The server device 100 includes a control section 201, a storage section 202, and a communication section 203 as a hardware configuration. The control unit 201 is, for example, a CPU (Central Processing Unit) or the like, and controls the entire server device 100 . The storage unit 202 is, for example, one of HDD (Hard Disk Drive), ROM (Read Only Memory), RAM (Random Access Memory), SSD (Solid State Drive), or any combination thereof, and stores programs and It stores data and the like used when the control unit 201 executes processing based on a program. The functions of the server device 100 are realized by the control unit 201 executing processing based on the program stored in the storage unit 202. The communication unit 203 is, for example, a NIC (Network Interface Card) or the like, connects the server device 100 to the network 150, and manages communication with other devices.

FIG. 3 is a diagram showing an example of the hardware configuration of the client device 110. The client device 110 includes a control section 301, a storage section 302, an imaging section 303, an input/output section 304, and a communication section 305 as a hardware configuration. The control unit 301 is, for example, a CPU, and controls the entire client device 110. The storage unit 302 is, for example, an HDD, ROM, RAM, SSD, or any combination thereof, and stores programs, data used by the control unit 301 when executing processing based on the program, and the like. . The functions of the client device 110 are realized by the control unit 301 executing processing based on the program stored in the storage unit 302. The imaging unit 303 is, for example, a camera or the like, and takes an image of a subject. The input/output unit 304 is, for example, a touch panel or the like, and inputs user operations to the control unit 301, displays processing results of the control unit 201, etc. The communication unit 305 connects the client device 110 to the network 150 and manages communication with other devices.

3. Description of probabilistic model, etc. (1) Introduction In this specification, U and L represent a set of tourists and tourist spots in tourist city c, respectively. Further, U's tourism record in city c is represented by D. FIG. 4 is a diagram showing the main symbology used in this specification.
The tourist spot l is indicated by coordinates (x _p , y _p ), where xp and yp represent corresponding latitude and longitude, respectively. The tourist spots L ₌ {l _{1 ,} . A tourist attraction is a particular coordinate that attracts tourists, and usually appears as a point or coordinate on a map.
The user experience is represented by (u, l, t, W). (u, l, t, W) is the action W (what did the user u (tourist) do) at the tourist spot l (where u visited) at time t (when did u visit)? represents. Words representing activities and scenes are used to describe users' actions at tourist destinations.
A tourist record D is collected from a tourist U in a given city c. The control unit 201 acquires the sightseeing record D using the geotagged image uploaded to the social image sharing site. Here, the geodagged image d is expressed as ( _ud , _ld , _td , _Wd ). It means that u _d took a photo representing the behavior of user d at time t _d and tourist spot l _d . The geotagged images (ie, tourist records) represent one or more user experiences. Therefore, D is a set D of tourist records d of a set U of tourists who visited tourist spot L in c and took pictures.
Tourist destination recommendation is a type of POI recommendation task, and when a tourist record D in c of a set of tourists U is given, tourist destination L={l ₁ , in city c for tourists u' _{. ..} , l _m } is a POI (Point of Interest) recommendation task. Here, u' may be a member of the tourist set U, but does not need to be a member.

(2) Preconditions Embodiment 1 focuses on recommending tourist spots based on what tourists want to experience at the tourist spot. For this purpose, we first assume the following (premises 1) to (premises 4) and propose a new user experience model for analyzing tourist behavior in destination city c.

(Premise 1)
Tourists may enjoy similar experiences and activities. This means you can group users based on their behavior or experience. Conversely, a user group based on experiences such as behavior and experience includes multiple tourists with similar preferences.

(Premise 2)
Each tourist destination provides multiple experiences (behaviors, actions, and experiences at the tourist destination), and similar tourist destinations may offer similar tourist experiences. In other words, tourist attractions can be grouped based on the user's experience. One experience can be enjoyed at multiple tourist spots.

(Premise 3)
Tourists enjoy a variety of experiences and perform different activities at different tourist destinations. Experience refers to a group of actions or actions that tourists can take at a tourist destination. For example, taking pictures, eating sweets, and meditating are common activities seen at tourist destinations. Behaviors can be grouped based on user experience.

(Premise 4)
User experience is influenced by the time tourists visit tourist attractions. That is, visit times are grouped based on user experience.

(3) User Experience Model Considering assumptions (1) to (4) and in order to analyze tourist behavior, the control unit 201 constructs a Basic User Experience Model (B-UEM), which is an example of a probabilistic model. . Here, tourists, tourist destinations, time, and behavior are used as observed variables, and user experience is used as a latent variable. FIG. 5 is a diagram showing an example of B-UEM. Further, FIG. 6 is a diagram showing an example of an algorithm by which the control unit 201 generates a probability model. The control unit 201 generates a user experience _gd from a category distribution having a parameter θ. The control unit 201 uses g _d to generate a tourist u _d , a sightseeing spot L _d , and a set of words w _{d,i (i=1, . . . , Nw)} related to g _d . Note that w _d,i indicates the i-th word among words indicating the behavior of a certain tourist in record d (∈D). In FIG. 5, white circles represent latent variables, and circles with a gray background represent observed variables.

In the model of this embodiment, the control unit 201 generates words for tourists, sightseeing spots, and actions from user experience, which is a latent factor. The four factors u, l, t, and W are conditionally independent. Each W is influenced only by user experience. This ensures that they are independent and have the same distribution without influencing each other.
The tourist record d is composed of the tourist u _d , the tourist destination visited l _d , the time t _d , and the action taken there w _d = {w _d,0 ,..., w _d,Nw } . These are generated from corresponding categorical distributions and expressed using discrete distributions. Tourists, tourist destinations, words, and time are generated through the potential user experience g _d . g _d is generated from a categorical probability distribution with parameter θ and is expressed as follows.

From (premise 1), users are generated from user experience. Therefore, the following formula holds.

Similarly, tourist spots are generated from the probability distribution φ _gd of tourist spots for each user experience based on (premise 2).

Further, the time period is generated from the probability distribution τ _gd for the user experience based on (premise 4).

In addition, let w _d,i be a word representing an action, and the fact that this variable stands alone means that there are multiple (N _w ) possible actions for a tourism record where the tourist and tourist destination are fixed. This shows that (premise 3). Generation of each word is performed as follows.

(4) Tourist destination recommendation using a pseudo evaluation mechanism The proposed B-UEM includes four aspects of the tourist experience (who: who, where: where, when: what). For some of these aspects, the missing information can be estimated using other aspects given. Here, without loss of generality, we will exemplify tourist destination recommendation using p(l|u) for a tourist u∈U with a known history. For a tourist spot l _i ∈L, the following holds true.

Therefore, p(l _i |u) is calculated as follows.

Similarly, actions can be recommended by estimating p(w _j |u) as follows.

The model of this embodiment cannot be directly applied to recommend tourism resources to a new user u' who has no tourism record. To address this issue, we propose a pseudo evaluation mechanism. Suppose that n images from other users' sightseeing records N (n<<N) are provided to a new user u', and u' evaluates the images. All users' ratings for all images
Under the assumption that u' follows a known distribution F _rating , it is possible to estimate u''s evaluation of tourist spots and activities, and use this to predict u''s preferred tourist spots and activities.
Evaluation value information
Expressed as
Here, l _k and w _k represent the position and action of image I _k , respectively. In the case of tourist spot l _i ∈L and action w _j ∈W, it is as follows.

Applying Bayes' theorem,

Under the assumption,

Here, rank (l _k , w _k , I _k ) is the rank of (l _k , w _k , I _k ) in N images. However, since the time cost to accurately calculate the distribution of rank (l _k , w _k , I _k ) is high, the Monte Carlo method is used to approximate the distribution to increase the speed.

In this way, tourist spots can be recommended according to various situations using two methods based on UEM.
・Tourist spots can be recommended to users who have historical data without using a pseudo mechanism.
・For new users who do not have historical data, it is possible to provide multiple images (or words expressing tourist spots or actions) to be evaluated and recommend tourist spots based on a pseudo evaluation mechanism and UEM. can.
In addition, if feedback for recommendation can be received, p(g|R) can be used as a prior probability instead of p(g) in equation (12) for the next recommendation, so the user's preferences can be more accurately determined. It becomes possible to make recommendations that reflect the current situation, thereby creating a user-adaptive recommendation system for new tourists.

(5) Other probability models Variations of the above-mentioned B-UEM will be explained. Two additional assumptions are added to the assumptions described above.
(Premise 5)
Behavior at a certain tourist destination is influenced by that tourist destination. In other words, words related to actions are generated by the context of the tourist destination.
(Premise 6)
Behavior at a certain tourist spot is influenced by time. That is, some words are generated based on the temporal context.

Based on these two assumptions, the Basic User Experience Model (B-UEM) is divided into three types: the Space-Time User Experience Model (ST-UEM), the Spatial User Experience Model (S-UEM), and the Temporal User Experience Model (T-UEM). There will be two variations. Three probabilistic models, ST-UEM, S-UEM, and T-UEM, are shown in FIGS. 7, 8, and 9. FIG. 7 is a diagram illustrating an example of ST-UEM. Compared to B-UEM, the action word w _d,i is influenced by the tourist spot l _d and the time t _d . ST-UEM considers all assumptions (premises 1 to 6). ST-UEM models variables λ as factors that most influence behavior, such as user experience, tourist destinations, and time. Generation of this variable depends on the tourist destination _ld since λ differs depending on the tourist destination. The variable λ _ld is used to generate w _d,i . In particular, after generating g _d and l _d , a variable λ _ld is generated for each tourist spot, and is treated as weighting for generating words that describe an action. In addition,
, and the first argument of the polynomial is the number of generation times.

FIG. 8 is a diagram showing an example of S-UEM. Compared to B-UEM, S-UEM does not have a time period t _d variable. FIG. 9 is a diagram showing an example of T-UEM. Compared to B-UEM, in T-UEM, the word w _d,i representing an action is influenced by the time period t _d .

4. Information Processing (1) Overview of Information Processing An overview of information processing executed by the control unit 201 of the server device 100 will be described.
The control unit 201 acquires a plurality of pieces of evaluation value information. Here, the plural pieces of evaluation value information are, for example, information indicating the user's preferences regarding tourist spots and the like. The control unit 201 determines the user's selection probability based on the acquired plural pieces of evaluation value information and the probability model. Here, the user's selection probability is the probability of a tourist spot being selected by the user as a destination from among a plurality of tourist spots.
By performing such processing by the control unit 201, it is possible to obtain a selection probability that reflects the user's preferences.

(2) Details of Information Processing FIG. 10 is an activity diagram showing an example of information processing in the server device 100. Here, the explanation will be based on the assumption that the user is a travel planner who is researching tourist experiences. However, this does not limit this embodiment.

When a user who is a travel planner accesses the URL (Uniform Resource Locator) of a service proposed by the server device 100, in A701, the control unit 201 generates a login screen 800 and controls the input/output of the client device 110. control section 304 to display the information. FIG. 11 is a diagram showing an example of a login screen 800. The login screen 800 includes an input field 801 for entering a user name, an input field 802 for entering a login password corresponding to the user name, a login button 803 that is selected when logging in, and a registration link 804 that is selected during user registration. and are included.

When the registration link 804 is selected, the control unit 201 determines in A702 that there is no user ID, and advances the process to A703. If the user name is entered in the input field 801, the login password is entered in the input field 802, and the login button 803 is selected, the control unit 201 advances the process to A706.
At A703, the control unit 201 receives user information via the account registration screen. The control unit 201 obtains the user name of the unregistered user, the password corresponding to the user name, etc. via the account registration screen. The acquired information is held and managed, for example, in the storage unit 202 or the like.

When the account registration is completed, in A704, the control unit 201 generates a questionnaire screen 900 regarding tourist spots, etc., and controls the screen to be displayed on the input/output unit 304 of the client device 110. FIG. 12 is a diagram showing an example of the questionnaire screen 900. The questionnaire screen 900 includes multiple images 901 and multiple sliders 902. The slider 902 is configured so that the degree of interest in the corresponding image 901 can be input. In this way, by configuring the system so that it is possible to input the level of interest, rather than simply choosing between likes and dislikes, it is possible to propose more accurate sightseeing experiences. Additionally, the questionnaire screen 900 includes a plurality of images and a slider corresponding to each of the plurality of images. More specifically, the questionnaire screen 900 in FIG. 9 includes four images 901a, 901b, 901c, and 901d, and sliders 902a, 902b, 902c, and 902d corresponding to each of the four images. By having a plurality of images 901 in this way, it is possible to increase the amount of preference information, and it is possible to realize a more accurate proposal for a sightseeing experience.
In A705, the control unit 201 receives evaluation value information of each image as a response to the questionnaire via the questionnaire screen 900. Here, the evaluation value information is as described above.
It is expressed as The process of A705 or the processes of A704 and A705 is an example of a process of acquiring a plurality of pieces of evaluation value information. Note that, as described above, the screen shown in FIG. 12 is displayed on the input/output unit 304 of the client device 110. That is, information indicating the user's tourism preferences is acquired from the client device 110. Here, the information indicating the user's tourism preferences is an example of evaluation value information. Further, the client device 110 is an example of a user's terminal device.

In A706, the control unit 201 converts the evaluation value information into a visit probability. To explain more specifically, the control unit 201 calculates the user's visit probability based on the plurality of acquired evaluation value information and the probability model. Conversion of evaluation value information into visit probability is performed using equations (10) to (13) described above. Visit probability is an example of selection probability.
In A707, the control unit 201 generates a recommendation screen regarding tourist spots based on the visit probability, and controls the screen to be displayed on the input/output unit 304 of the client device 110. The process A707 is an example of a process that outputs information on a plurality of tourist spots recommended to the user based on the selection probability. FIG. 13 is a diagram showing an example of the recommendation screen 1010. The example in FIG. 13 includes a plurality of sightseeing experiences 1020 sorted in descending order of the probability that the user will visit them. Further, in the example of FIG. 13, a list of action words 1022 indicating actions included in the sightseeing experience is included.

More specifically, in the example of FIG. 13, on the left side of the recommendation screen 1010, the trademark temple, the patent history museum, and the design shrine are included as sightseeing spots 1021 included in the sightseeing experience 1020 for the user. Furthermore, a plurality of action words 1022 representing desirable actions at each tourist spot 1021 are listed. For example, if there is a large bell at Trademark Temple and it is desirable to take a tourist photo in front of the bell, action words 1022 such as "bell" and "photographing" are written. In addition, if it is possible to arrange zazen at a trademark temple, and even if this zazen is not a typical sightseeing activity, it is judged to be desirable for the user or the group to which the user belongs, the action word ``zazen'' 1022 is described. Similarly, action words 1022 indicating desirable actions are proposed and written at the Patent History Museum and Design Shrine.

Further, the recommendation screen 1010 may include map information 1030 representing tourist spots included in the sightseeing experience 1020. In the example of FIG. 13, map information 1030 is displayed on the right side of the recommendation screen 1010, and the location of the tourist spot 1021 is indicated on the map information 1030 by a spot pin icon 1040 so that the user can understand it. Here, an icon is something that is expressed as a symbol using a simple picture. The on-screen position of the recommendation screen 1010 that proposes the sightseeing experience 1020 and the on-screen position of the recommendation screen 1010 of the map information 1030 are merely examples, and are not limited to this. The sightseeing experience 1020 and the map information 1030 may be arranged above and below the recommendation screen 1010, or may be arranged with the left and right reversed.

In A708, if a button or the like to end the display is selected, the control unit 201 ends the process shown in FIG. 10.
Note that the control unit 201 may collect user feedback regarding the recommended sightseeing experiences and update the user preference analysis.
Further, in A702, if the control unit 201 determines that there is a user ID, it recommends tourist spots based on the user's past usage history and the like.

As described above, according to the process of this embodiment, it is possible to recommend appropriate sightseeing spots etc. to both users without historical data and the like and to users with historical data etc. Further, this recommendation can be applied whether it is advance recommendation or real-time recommendation.

<Modified example 1>
Modification 1 of Embodiment 1 will be described. In the embodiment described above, POI recommendation, particularly tourist POI recommendation, was explained as an example. However, the above-described processing performed by the control unit 201 can also be extended to product recommendation, product advertisement purchase prediction, click prediction, etc.
That is, the control unit 201 of Modification 1 acquires evaluation value information of a product being sold as evaluation value information. More specifically, the control unit 201 acquires evaluation value information of a product from a website related to product sales. Product evaluation value information may be, for example, word-of-mouth information written by users who have purchased the product, or information indicating how many points out of five points the product is given. The control unit 201 determines the user's selection probability regarding product purchase based on the acquired plurality of evaluation value information and the probability model. The control unit 201 outputs information regarding a plurality of products purchased by the user based on the determined selection probability. Here, an example of information related to a plurality of products related to a user's purchase is information on products recommended to the user. In this case, the information is output to the input/output unit 304 of the client device 110 of the corresponding user. Another example of information related to a plurality of products related to a user's purchase is information related to a prediction that a user will select a purchase button for a product. In this case, the information output destination is a display of a predetermined PC or the like.
According to modification example 1, recommended products are presented to the user on a website that sells products, or products that are likely to be purchased by the user or products that are likely to be clicked are presented on the website, etc. You can do it.

<Modified example 2>
Modification 1 was explained using a product as an example. However, products may also be read as services.
That is, the control unit 201 of the second modification obtains evaluation value information of the provided service as evaluation value information. More specifically, the control unit 201 acquires service evaluation value information from a website related to service sales. The evaluation value information of a service may be, for example, information written by a user who received the service, or information indicating a score out of five. The control unit 201 determines the user's selection probability regarding the use of the service based on the acquired plural pieces of evaluation value information and the probability model. The control unit 201 outputs information regarding a plurality of services related to the user's use based on the determined selection probability. Here, an example of information related to a plurality of services related to use by the user is information on services recommended to the user. In this case, the information is output to the input/output unit 304 of the client device 110 of the corresponding user. Another example of information related to a plurality of services related to use by a user is information related to a prediction that a user will select a service use button. In this case, the information output destination is a display of a predetermined PC or the like.
According to modification example 2, a website that provides services presents recommended services to the user, or presents services that are likely to be used by the user or services that are likely to be clicked on the website, etc. You can do it.

<Additional notes>
The invention may be provided in each of the following aspects.
In the information processing device, the evaluation value information is information indicating the user's preferences regarding tourism, and the control unit determines the user's selection probability regarding the tourist destination based on the acquired plurality of evaluation value information and a probability model. An information processing device that searches for.
In the information processing device, the control unit acquires information indicating the user's preferences regarding tourism from the user's terminal device.
In the information processing device, the control unit outputs information on a plurality of tourist spots recommended to the user based on the selection probability.
In the information processing device, the evaluation value information is evaluation value information of a product being sold, and the control unit controls the user's selection regarding purchase of the product based on the acquired plurality of evaluation value information and a probability model. An information processing device that calculates probability.
In the information processing device, the control unit obtains evaluation value information of the product from a website related to product sales.
In the information processing device, the control unit outputs information regarding a plurality of products purchased by the user based on the selection probability.
In the information processing device, the information related to a plurality of products purchased by the user is information about products recommended to the user.
In the information processing device, the information related to the plurality of products purchased by the user is information related to a prediction that the user will select a purchase button for a product.
In the information processing device, the evaluation value information is evaluation value information of a service being provided, and the control unit controls the user's selection regarding the use of the service based on the acquired plurality of evaluation value information and a probability model. An information processing device that calculates probability.
In the information processing apparatus, the control unit obtains evaluation value information of the service from a website related to sales of the service.
In the information processing apparatus, the control unit outputs information regarding a plurality of services used by the user based on the selection probability.
In the information processing apparatus, the information related to the plurality of services used by the user is information on services recommended to the user.
In the information processing device, the information related to the plurality of services related to use by the user is information related to prediction that the user will select a service use button.
An information processing method executed by an information processing device, the method comprising: acquiring a plurality of pieces of evaluation value information; and determining a user's selection probability based on the plurality of acquired pieces of evaluation value information and a probability model.
A program for causing a computer to function as a control unit of the information processing device.
Of course, this is not the case.

For example, it may be provided as a computer-readable non-transitory storage medium that stores the above program.
Furthermore, the embodiments and modifications described above may be implemented in any combination.

Finally, although various embodiments according to the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. The new embodiment can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. The embodiments and their modifications are included within the scope and gist of the invention, as well as within the scope of the invention described in the claims and its equivalents.

100: Server device 110: Client device 150: Network 201: Control unit 202: Storage unit 203: Communication unit 301: Control unit 302: Storage unit 303: Imaging unit 304: Input/output unit 305: Communication unit 800: Login screen 801: Input field 802 : Input field 803 : Login button 804 : Registration link 900 : Survey screen 901 : Image 901a : Image 901b : Image 901c : Image 901d : Image 902 : Slider 902a : Slider 902b : Slider 902c : Slider 902d : Slider 1000 : Information processing system 1010 : Recommendation screen 1020 : Tourist experience 1021 : Tourist spot 1022 : Action word 1030 : Map information 1040 : Icon

Claims (16)

An information processing device,
has a control section,
The control unit includes:
Obtain multiple evaluation value information,
determining the user's selection probability based on the plurality of acquired evaluation value information and the probability model;
Information processing device. The information processing device according to claim 1,
The evaluation value information is information indicating the user's preferences regarding tourism,
The control unit includes:
determining the user's selection probability regarding the tourist destination based on the plurality of acquired evaluation value information and the probability model;
Information processing device. The information processing device according to claim 2,
The control unit includes:
acquiring information indicating the user's tourism preferences from the user's terminal device;
Information processing device. In the information processing device according to claim 2 or 3,
The control unit includes:
outputting information on a plurality of tourist spots recommended to the user based on the selection probability;
Information processing device. In the information processing device according to claim 1 or claim 2,
The evaluation value information is evaluation value information of a product being sold,
The control unit includes:
determining a user's selection probability regarding product purchase based on the plurality of acquired evaluation value information and a probability model;
Information processing device. The information processing device according to claim 5,
The control unit includes:
obtaining evaluation value information of the product from a website related to product sales;
Information processing device. In the information processing device according to claim 5 or 6,
The control unit includes:
outputting information regarding a plurality of products purchased by the user based on the selection probability;
Information processing device. The information processing device according to claim 7,
The information related to the plurality of products purchased by the user is information on products recommended to the user;
Information processing device. The information processing device according to claim 7,
The information related to the plurality of products purchased by the user is information related to a prediction that the user will select a purchase button for the product;
Information processing device. In the information processing device according to claim 1 or claim 2,
The evaluation value information is evaluation value information of the provided service,
The control unit includes:
determining the user's selection probability regarding the use of the service based on the acquired plurality of evaluation value information and the probability model;
Information processing device. The information processing device according to claim 10,
The control unit includes:
acquiring evaluation value information of the service from a website related to sales of the service;
Information processing device. The information processing device according to claim 10 or 11,
The control unit includes:
outputting information regarding a plurality of services related to use by the user based on the selection probability;
Information processing device. The information processing device according to claim 12,
The information related to the plurality of services used by the user is information on services recommended to the user;
Information processing device. The information processing device according to claim 12,
The information related to the plurality of services related to the user's use is information related to a prediction that the user will select a service use button,
Information processing device. An information processing method executed by an information processing device, the method comprising:
Obtain multiple evaluation value information,
determining the user's selection probability based on the plurality of acquired evaluation value information and the probability model;
Information processing method. A program for causing a computer to function as a control unit of an information processing apparatus according to any one of claims 1 to 14.