JP2010186283A - Information processor, information processing method, and information processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processor, an information processing method and an information processing program for extracting scenario candidates showing potential needs of a user. <P>SOLUTION: A behavior process of a user is modeled by using scenario configuration definition information, scenario sequence definition information, and scenario condition definition information. A group of questions to the user is calculated by referring to the user scenario history information of the user, the scenario information of the model to which the user belongs, and question definition information (step 1322). One question is extracted from the group of questions, and the question is executed (step 1324). Scenario components that can be newly extracted are calculated from the answer of the question by the user (step 1325). Scenario candidates showing the potential needs of the user are calculated by applying the condition/inference logic of the scenario condition definition information with respect to the user scenario history information and the calculated scenario components (step 1326). <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an information processing apparatus, an information processing method, and an information processing program for predicting a user's latent needs according to a user's action type and action process, and in particular, used by a person in charge at a consultation service at a sales office. The present invention relates to an information processing apparatus, an information processing method, and an information processing program that are suitable for use in a window terminal capable of predicting a user's latent needs in consultation with a user and proposing a solution.

  There is a technique called “recommendation” that predicts a user's desired purchase item based on the user's past purchase history and presents the product to the user. For example, in an electronic shopping site on the Internet, depending on the product page viewed by the user, multiple product information can be presented in the form of "the person who bought this product also bought this product" It presents information on what percentage of other people are buying what in the form of "What do you buy after seeing this product?" Further, when the user logs in to the electronic shopping site and the user's past purchase history is known, a product highly relevant to the user's past purchased product is presented in the form of “recommended product”. With such a recommendation function, it is possible to support the user's product purchase and to expand the product sales by predicting the purchase needs of the user.

  In particular, there is Patent Document 1 as a conventional technique related to the recommendation function. In patent document 1, it aims at implement | achieving the recommendation adapted to a user, reducing a user's burden. For this reason, Patent Document 1 relates to a method of providing preference information used for selection of information suitable for a user to a recommendation means by a computer. In Patent Document 1, the user is prompted to set preference information, and when preference analysis information is generated by analysis based on the user's activity history information, the preference analysis information is presented, and the preference information is set. The recommendation information set by the prompted user is provided to the recommendation means.

  In other words, in Patent Document 1, when the user analyzes his / her activity history information and obtains preference analysis information, the user sets his / her preference information with reference to the preference analysis information. For this reason, in patent document 1, the burden of the user's preference information setting work can be reduced. Further, the user can set preference information reflecting the current preference. And the precision of recommendation can be improved.

JP 2003-271853 A

  In such a conventional method, a product that a user seeks is predicted with reference to another user's purchase history similar to the user's past purchase history. By analyzing behaviors such as product browsing and product purchases and the behavior of other similar users, it is possible to identify the essential background of the user's intention to take such behavior and capture potential needs. It is difficult. In particular, Patent Document 1 “statistically analyzes patterns of items purchased by many users in the past (EC sites) and items accessed by many users in the past”. I ca n’t find the right background. In addition, an active operation of the user (user) is required.

  The object of the present invention is to use the user's behavioral process, model the user's behavior type, and clarify the user's behavior scenario, thereby deriving from the user's superficial behavior It is to provide an information processing apparatus, an information processing method, and an information processing program capable of capturing potential usage needs in consideration of the essential background of user behavior instead of needs.

  In order to achieve this object, in the present invention, scenario configuration definition information that defines components of a “scenario” that indicates a usage scene in a certain aspect of the user, with the behavior type and behavior process of the user as “persona”. , Scenario order definition information that defines relationships such as the occurrence order of scenario components, and scenario condition definition information that defines constraints and inference conditions related to scenarios as formal logic (logic). Perform processing including presentation. In the present invention, a scenario is generated by modeling a user's action process using scenario configuration definition information, scenario order definition information, and scenario condition definition information, and the user's action (including answer to a question) is generated from this scenario. It can also be expressed as generating a new scenario based on. More specifically, a new scenario based on this scenario is dynamically generated based on the user's answer to the question based on this scenario and each information described above. The present invention includes the creation of this new scenario by changing the scenario for the question, for example, by changing the components of the scenario, or extracting other scenarios that already exist (created). Note that the generation of a scenario before a new scenario also includes a change to a scenario (candidate) generated (defined) in advance and extraction from a scenario (candidate) that already exists (generated). In addition, the change of the scenario component includes replacement, addition, and a combination thereof.

The above processing can also be expressed as follows. Predefine multiple personas according to user behavior types, define the behavior process for each persona,
Assign appropriate personas to users in advance. And, the user scenario history information that records the history of the scenario that the user has acted so far, the above three definition information about the persona to which the user belongs, and the question contents and the conditions regarding the question are defined according to the persona and scenario A set of questions that can be performed on the user is calculated using the question definition information. One optimal question to be executed next is calculated from the set of questions and the user is asked a question. By referring to the scenario element definition information that defines the scenario components, the scenario components that indicate the user's needs with respect to the user's behavior types and behavior processes are newly added by answering the above question by the user. If it is possible to calculate, the constituent elements of the scenario derived by the calculation are retained. Whether or not another scenario can be calculated by applying the condition / inference logic defined in the scenario condition definition information to the user scenario history information and the retained scenario component calculate. If another new scenario can be calculated, it is determined that there is a possibility of a scenario indicating the potential needs of the user, and is added as a scenario candidate.

  If no persona is associated with the user, use information about any persona scenario, information about another user's scenario, or information about an anonymous scenario where the user cannot be identified as alternative information. Calculate user scenario candidates.

  In addition, in this invention, both the aspect which implement | achieves the above-mentioned process with what is called a server, and the aspect implement | achieved with a terminal are contained. Further, a program for causing these information processing apparatuses to execute processing is also included.

  According to the present invention, it is possible to capture more potential needs.

It is a figure which shows the function structural example of the information processing system in embodiment of this invention. It is a figure which shows an example of the utilization log information of embodiment. It is a figure which shows an example of the scenario element definition information of embodiment. It is a figure which shows an example of the scenario structure definition information of embodiment. It is a figure which shows an example of the scenario order definition information of embodiment. It is a figure which shows an example of the scenario condition definition information of embodiment. It is a figure which shows an example of the user corresponding information according to persona of embodiment. It is a figure which shows an example of the user scenario log | history information of embodiment. It is a figure which shows an example of the question definition information report of embodiment. It is a figure which shows the whole processing flow which calculates a user's scenario candidate in the server of embodiment. It is a figure which shows an example of the user information screen in the window terminal of embodiment. It is a figure which shows an example of the schematic image of the scenario information of the persona of embodiment. It is a figure which shows an example of the schematic image of the scenario information of the user of embodiment. It is a figure which shows the sub-processing flow which calculates a user's scenario candidate in the server of embodiment. It is a figure which shows an example of the hearing assistance screen to the user in the window terminal of embodiment. It is a figure which shows an example of the schematic image of the scenario candidate of the user of embodiment. It is a figure which shows an example of the proposal assistance screen to the user in the window terminal of embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
In this embodiment, at a bank sales office consultation desk, a bank employee (person in charge of the consultation desk) finds the essential background of the user's needs in communication with the user from the viewpoint of the user, and the user. An example of an information processing apparatus, an information processing method, and an information processing program for making a proposal that captures potential needs is used. However, the technical idea of the present invention is not limited to this example. The potential needs may be those that the user is conscious or unconscious.

  FIG. 1 is a diagram illustrating a functional configuration example of an information processing system according to the present embodiment. The plurality of user terminals 2, the plurality of window terminals 3, and the server 1 are connected via a communication network 4.

  As shown in FIG. 1, the user terminal 2 has four functions, the window terminal 3 has four functions, and the server 1 has four functions, and the hardware and hardware of each device are controlled. These functions are activated by.

  The first function of the user terminal 2 uses an input device such as a mouse, a keyboard, and a touch panel, and refers to an application running on the user terminal 2, for example, a WWW (World Wide Web) page on the Internet. This is an input function 21 for performing operations such as input and selection on the WWW browser.

  The second function of the user terminal 2 includes operation information such as mouse operation, keyboard operation, and touch panel operation performed by the user using the input function 21, and operation information such as reference, input, and selection in an application such as a WWW browser. This is a usage history acquisition function 22 for acquiring as a usage history.

  Note that some operation information in the application, for example, page information referred to by the WWW browser, transaction information executed on the WWW browser, and the like can be acquired as a usage history by the usage history management function 12 of the server 1. .

  The third function of the user terminal 2 is a transmission / reception function that transmits the usage history acquired by the usage history acquisition function 22 to the server 1 and receives information related to screen display such as an application screen such as a WWW page from the server 1. 23.

  A fourth function of the user terminal 2 is a screen display function 24 for performing screen display based on screen information by an application such as a WWW browser.

  The first function of the window terminal 3 is an input function 31 for performing operations such as input and selection for applications running on the window terminal 3 using input devices such as a mouse, a keyboard, and a touch panel. It is.

  The second function of the window terminal 3 is a navigation function 32 for supporting the consultation service of the consultation window staff in communication between the user and the consultation window staff. For example, the person in charge of the consultation desk presents questions to the person in charge and asks the person in charge the next question that is appropriate for that answer according to the user's answer. Later, the potential needs of users are predicted and presented to the person in charge.

  Note that the process of calculating the question to be presented according to the user's answer and the process of predicting and presenting the potential needs of the user are performed by the scenario calculation function 13 of the server 1 and the navigation function of the window terminal 3 32 operates as a client side program. However, all or part of the scenario calculation function 13 that operates as a server-side program may be distributedly processed as a client-side program.

  The third function of the window terminal 3 is to send operation information generated in the client side program operating as the navigation function 32 to the server 1 and from the server 1 to the client side program operating as the navigation function 32. The transmission / reception function 33 receives command information, screen information, and the like.

  A fourth function of the window terminal 3 is a screen display function 34 for performing screen display based on screen information by a client-side program operating as the navigation function 32.

  The first function of the server 1 receives the usage history acquired by the usage history acquisition function 22 from the user terminal 2, transmits information related to screen display such as a WWW page or an application screen to the user terminal 2, and the window terminal 3 for receiving operation information generated in the client-side program operating as the navigation function 32 and transmitting / receiving command information, screen information, etc. to the client-side program operating as the navigation function 32 of the window terminal 3 Function 11.

The second function of the server 1 is a usage history management function for collecting and managing the usage history acquired by the usage history acquisition function 22 of the usage terminal 2 and the usage history (transaction contents, transaction results, etc.) acquired in the server 1. 12.
A third function of the server 1 is a scenario calculation function 13 for calculating a user's potential needs as scenario candidates in communication between a person in charge using the navigation function 32 of the window terminal 3 and the user. It is. Details of the scenario calculation function 13 will be separately described later.

  A fourth function of the server 1 is a storage function 14 that manages information handled in the server 1. The storage function 14 includes usage history information 141, scenario element definition information 142, scenario configuration definition information 143, scenario order definition information 144, scenario condition definition information 145, user correspondence information by persona 146, user scenario history information 147, questions Definition information 148 is stored.

  Note that any two or all of the server 1, the user terminal 2, and the developer terminal 3 are physically one device not via the communication network 3, and the functions described above are provided as one device. You may have in.

  Hereinafter, usage history information 141 managed by the storage function 14 of the server 1, scenario element definition information 142, scenario configuration definition information 143, scenario order definition information 144, scenario condition definition information 145, user correspondence information by persona 146, The user scenario history information 147 and the question definition information 148 will be described.

FIG. 2 is a diagram illustrating an example of the usage history information 141 according to the embodiment.
In FIG. 2, the operation time 1411 is the time when the usage history is acquired by the usage history acquisition function 22 of the user terminal 2 or the usage history management function 12 of the server 1, the user ID 1412 is an ID for identifying the user who performed the operation, The operation type 1413 is the type of operation by the user, and the operation content 1414 is the content of the operation by the user.

  The operation type 1413 is identified as “reference” when the user refers to the WWW page, and “operation” when an execution operation such as a contract or search is performed on the transaction screen or search screen of the application. There may be operation types other than “reference” and “operation”.

  The operation content 1414 includes a URL (Uniform Resource Locator) of the referred WWW page when the operation type 1413 is “reference”, and an identifier that can identify the operation executed in the application on the server side when the operation type 1413 is “operation”, or execution Program name.

  The operation type 1413 and the operation content 1414 can be acquired by the log creation function of the WWW server and the log creation function of the server-side program. In addition, by using the existing Ajax (Asynchronous JavaScript + XML) technology or Adobe Flash (R) technology, it is possible to acquire detailed operation information of the user's mouse operation and keyboard operation (these Each name is a trademark).

  For example, the record 1415 indicates that “the user whose user ID is 1001 referred to the investment trust sale page (fund-cell.html) at 15:05:10 on December 1, 2008”. Further, the record 1416 indicates that “a user with a user ID of 1001 executed a sell-simulation simulation at 15:08:15 on December 1, 2008”.

  The usage history information 141 shows the usage history acquired by the WWW server and the usage history acquired by each of the plurality of applications as one piece of information, but may be managed (stored) separately as a plurality of information. .

  In addition to the usage history that can be acquired by the user terminal 2 or the server 1, the usage history may be acquired from an affiliated external system or external terminal such as an SNS (Social Networking Service) site or a community site.

  In the following, the “scenario element” is defined as the individual minimum action unit of the user. For example, the user's “Investment Trust Sale Check”, “Investment Trust Price Check”, “Investment Trust Cancellation Simulation”, “Home Loan Contract”, and the like. Details of the definition of the scenario element will be described later with reference to FIG.

  In the following, a “scenario” is defined as a meaningful group of actions composed of a plurality of scenario elements. For example, a user's “mortgage contract by sale of investment trust”, “asset management using bond-backed loan”, and the like.

Below, the persona method is used as a method for classifying the user's action type.
The persona method is to create a fictitious user model (virtual human figure with personality) by setting the user's profile in detail, product development, promotion / promotion for that "person (persona)", This is a method for pursuing needs. The “persona” is information indicating a “virtual person image having a personality”. Although it is virtual, it may be information indicating a real image of the user.

  In the present embodiment, a plurality of persona behavior processes are defined in advance by utilizing the scenario defined above. Details of the scenario and the modeling of the user using the persona method will be described with reference to FIGS.

  Below, a persona is assigned in advance to each user. However, there may be a user who cannot be assigned to a persona because information about the user (behavioral process information, that is, scenario information) is insufficient. Details of management of users belonging to the persona will be described later with reference to FIG.

  Hereinafter, the “scenario information” of the user is defined as a scenario element arranged in time series according to the actual behavior process of the user. Persona scenario information is defined in the same way as for users. The scenario information of each user is managed in the storage function 14 of the server 1. Details of the scenario information of each user will be described later with reference to FIG.

FIG. 3 is a diagram illustrating an example of the scenario element definition information 142.
In FIG. 3, a scenario element ID 1421 is an ID for uniquely identifying each scenario element, a scenario element name 1422 is a name of the scenario element, an operation type 1423 is a type of operation by the user, an operation content 1424 is a content of operation by the user, The satisfaction condition 1425 is a condition for the operation content to be satisfied as a scenario element.
One scenario element is composed of a plurality of sets (operation type, operation content, fulfillment condition).

  The satisfaction condition 1425 indicates a condition to be satisfied, such as the generation order, the number of occurrences, and the generation period of the corresponding operation content in order for the scenario element to be satisfied. In addition, this condition may describe the establishment condition over a plurality of operation contents. However, it is not always necessary to describe all constraints such as the order of occurrence, the number of occurrences, and the duration of occurrence (it is sufficient to include at least one). The scenario element definition information 142 is defined in advance and stored in a storage device. For example, the record 1426 indicates that “in a user's usage history, the account balance page (account.html) is referenced first, the investment trust price page (fund-price.html) is referenced second, the third If there is a history such as execution of investment trust sale simulation (sell-simulation), a scenario element called investment trust sale check is established ”. The scenario element definition information 142 in FIG. 3 may be defined for each persona.

  Next, in order to model a behavior process using a persona, scenario configuration definition information 143, scenario order definition information 144, and scenario condition definition information 145 are defined, and these definition information are used in combination.

FIG. 4 is a diagram illustrating an example of the scenario configuration definition information 143.
In FIG. 4, a persona ID 1431 is an ID that uniquely identifies each persona, a scenario ID 1432 is an ID that uniquely identifies a scenario defined as a persona action process, a scenario name 1433 is the name of the scenario, and a scenario element ID 1434 is the scenario. The IDs and weights 1435 of the individual scenario elements constituting the scenario are values indicating the importance of the scenario elements in the scenario. Attributes other than those described above may be added to the scenario configuration definition information 143 column. The scenario configuration definition information 143 is defined in advance and stored in a storage device.

  For example, in FIG. 4, the persona ID of a certain persona is P1, and this persona is composed of a plurality of scenarios (S0, S1, S2,...). Further, the scenario S1 is composed of a plurality of scenario elements (a1, a2, a3, a6,...).

  In this case, in scenario S1, the weight of scenario element a1 is 20, the weight of scenario elements a2 and a3 is 10, the weight of scenario element a6 is 30, and the weight of scenario element a6 is the largest in scenario S1, This indicates that this is a key scenario (important action) in scenario S1. The scenario configuration definition information 143 defines the scenario that configures the persona, the scenario elements that configure the scenario, the weights of the scenario elements, and the like. Logics (programs) for characterizing behavioral processes such as relevance such as the order relation of scenario elements in a scenario, constraints between scenarios and between scenario elements, and establishment conditions are scenario order definition information 144 and scenario condition definition information, respectively. 145.

FIG. 5 is a diagram illustrating an example of the scenario order definition information 144.
In FIG. 5, persona ID 1441 is an ID that uniquely identifies each persona, scenario ID 1442 is an ID that uniquely identifies a scenario defined as a persona action process, and order 1443 is an order relation between scenarios or order relations between scenario elements. Multiple definitions. Attributes other than those described above may be added to the scenario order definition information 144 column. The scenario order definition information 144 is defined in advance and stored in the storage device.

  When defining the order relationship between scenarios in a persona, “-” is described in the scenario ID 1442 column of the persona, and the order relationship between scenarios is specified in the order 1443 column. Here, “-” indicates information indicating the order of scenarios, and may be defined in another form. In the example of FIG. 5, the third record in FIG. 5 indicates that “the scenario S1 occurs after the scenario S0 (S0 → S1) with respect to the scenario defined in the persona P1. This indicates that scenario S2 occurs after S0 (S0 → S2) ”. This example is a case (information indicating) that there is a possibility of branching from scenario S0 to scenario S1 or S2.

  When defining an order relationship between scenario elements in a scenario, the scenario ID is designated in the scenario ID 1442 field of the persona, and the order relationship between scenario elements is designated in the order 1443 column.

  For example, the first record in FIG. 5 indicates that “the scenario element a2 occurs after the scenario element a1 (a1 → a2) and the scenario element a2 follows the scenario element defined in the scenario S1. Indicates that scenario element a3 occurs (a2 → a3) ”. Note that it is not necessary that all scenarios or scenario elements defined in the scenario configuration definition information 143 are defined as an order relationship. It is sufficient that a scenario including at least one scenario element can be defined.

FIG. 6 is a diagram illustrating an example of the scenario condition definition information 145.
In FIG. 6, a persona ID 1451 is an ID that uniquely identifies each persona, a condition ID 1452 is an ID that uniquely identifies a condition / inference 1453 described later for the persona, and a condition / inference 1453 is a condition that must be satisfied for the persona. Or it describes the results obtained by inference.

  Attributes other than those described above may be added to the scenario condition definition information 145 column. The scenario condition definition information 145 is defined in advance and stored in the storage device.

  For example, the first record in FIG. 6 (the record whose condition ID is C1) is “the scenario element coverage in S1 needs to be 60% or more to satisfy that the scenario is S1”. Indicates conditions. In addition, the second record in FIG. 6 (the record whose condition ID is C2) is inferred that “when scenario element a6 exists in scenario S1, scenario element b2 also exists (infers that scenario element b2 exists in scenario S2)”. Results are shown.

  In the example of FIG. 6, the description of the condition 1453 uses a natural language, but on implementation, a logical expression such as a conditional statement, a repeated statement, a comparison, and an assignment is formally described using a programming language. The conditions and inference 1453 described in the scenario condition definition information 145 are top-down by a quantitative survey and a qualitative survey, in addition to scenarios that can be reflected in the persona bottom-up from the user's past usage history and current behavior process. Describe the scenario you want to draw as a persona and the know-how of the consultant and bank planning department as logic.

  Therefore, by describing the behavior process as logic using the scenario condition definition information 145 in addition to the scenario configuration definition information 143 and the scenario order definition information 144, it is possible to derive the potential needs of the user.

FIG. 7 is a diagram illustrating an example of person-by-persona user correspondence information 146.
In FIG. 7, a persona ID 1461 describes an ID for uniquely identifying each persona, and a user ID 1462 describes a plurality of user IDs belonging to the persona. For example, the first record in FIG. 7 (a record whose persona ID is P1) indicates that “user IDs belonging to persona P1 are 1001, 1002,...”. It should be noted that the persona to which the user belongs is determined in advance, and this determination to be registered may be made by a human (system operator or the like) or may be automatically determined systematically. . However, there may be users who cannot determine which persona they belong to because there is no persona that can be associated with the user, or the persona cannot be determined due to insufficient information about the user. In this case, it is not necessary to register a persona to be associated (blank), or standard data called a standard persona may be registered.

FIG. 8 is a diagram illustrating an example of the user scenario history information 147.
In FIG. 8, a user ID 1471 is an ID for identifying a user, a time 1472 is a time when a scenario element of the user is generated, and a scenario element ID 1473 is an ID of a scenario element generated by the user. Attributes other than those described above may be added to the user scenario history information 147 column. The time 1472 is, for example, the time when the scenario element first occurs. Referring to the usage history information 141 in FIG. 2 and the scenario element definition information in FIG. 3, the occurrence time of the scenario element a1 is the time when the first operation content (refer to account.html) is first recorded “2008/12 / 01 15:00:01 ”.

  The user scenario history information 147 refers to the scenario element definition information 142 in FIG. 3 based on the use history information 141 obtained by the use history acquisition function 22 of the user terminal 2 and the like. Are extracted and added to the user scenario history information 147.

  When scenario elements are extracted from the usage history information 141 of the user with reference to the scenario element fixed information 142, the operation content 1424 does not necessarily appear as a series of consecutive records vertically adjacent to each other in the usage history information 141. do not have to. That is, other operation contents may be included in the usage history information 141.

FIG. 9 is a diagram illustrating an example of the question definition information 148.
In FIG. 9, persona ID 1481 is an ID that uniquely identifies each persona, scenario ID 1482 is an ID that uniquely identifies a scenario in the persona, question ID 1483 is an ID that uniquely identifies a question in the scenario, and question 1484 is in the scenario. The question contents and answer options and the condition 1485 are constraints for asking the question. The question definition information 148 is defined and registered in advance. In addition, what is “predefined” including this information may be stored manually input by an administrator or the like. The condition 1485 describes a condition related to the order in which questions are asked and a condition that is effective when a certain condition is satisfied.

  The question definition information 148 is used when the consultation counter person in charge asks the user in the navigation function 32 of the window terminal 3 in order to calculate the scenario candidate of the user in the scenario calculation function 13 of the server 1. It is. As for the question, the next question (or a set of questions) is dynamically calculated according to the persona to which the user belongs, the scenario candidate estimated by the user, the content of the user's answer and the condition 1485 related to the question. .

  Define two people who have similar (especially the same) scenarios in a scene, but ask different questions appropriate to the scenario of each persona if the personas they belong to are different be able to.

  By repeating such a question several times, a scenario element can be newly acquired, and as a result, it can be used for calculation of a scenario candidate.

Below, the process flow of the scenario calculation function 13 of the server 1 is demonstrated.
In the consultation service with the user, the person in charge of the consultation window asks the user the question by using the navigation function 32 of the window terminal 3, and the user scenario is clarified so that the potential of the user can be determined. Dig up your needs.

  The scenario calculation function 13 is used by the server 1 in the process of clarifying the scenario.

FIG. 10 is a diagram illustrating an overall processing flow for calculating user scenario candidates in the server 1 of the embodiment.
In step 131, it is checked whether or not a persona is associated with the user by referring to the persona-specific user correspondence information 146. If the persona is associated with the persona, the process proceeds to step 132 and associated with the persona. If not, the process proceeds to step 136.

  A user ID for identifying the user is searched from the persona-specific user correspondence information 146, and if there is a corresponding persona ID, the persona is associated with the persona. In the subsequent processing flow, information corresponding to this persona ID is obtained. Refer to

In step 132, a scenario candidate corresponding to the user is calculated while asking the user a question.
Details of this step will be described later.

  FIG. 11 is a diagram illustrating an example of a user information screen in the window terminal 3 according to the embodiment. When the persona is associated with the user, information related to the user as shown in FIG.

  The screen area 601 displays information about a persona associated with the user (for example, an outline of the persona). For example, as shown in FIG. 11, the persona type is “active management group”, and those belonging to this type have the characteristics of “utilizing IT” and “active to asset management”. indicate. Also, photos that make you imagine this type of person will be posted.

  A screen area 602 displays basic customer information regarding the user. For example, as shown in FIG. 11, the user name, annual income, and transaction information (account opening status, transaction asset balance, etc.) are displayed. As shown in FIG. 11, the screen area 603 displays information related to the current state of the user, such as reservations for visiting a consultation counter.

  As shown in FIG. 11, the screen area 604 displays history information related to the user's past transactions and detailed operations (for example, conducting investment trust cancellation simulation by Internet banking). The screen area 605 displays guidance (advice) that supports customer service for the consultation service representative.

FIG. 12 is a diagram illustrating an example of a schematic image of persona scenario information according to the embodiment.
From the scenario configuration definition information 143 and the scenario order definition information 144, the persona whose persona ID is P1 includes scenarios S0, S1, S2, and the like, and there is an order relationship such as S0 → S1, S0 → S2.

  Further, from scenario configuration definition information 143 and scenario order definition information 144, scenario elements a1, a2, a3, a6, etc. exist in scenario S1 of persona P1, and a1 → a2, a2 → a3, a3 → a6, etc. An order relationship exists.

  FIG. 12 shows an order relationship between scenarios of a persona whose persona ID is P1 and an order relationship between scenario elements in the scenario. An area 501 is an order relationship between the scenario elements in the scenario S0, and the area. Reference numeral 502 denotes an order relationship between scenario elements in the scenario S1, and an area 503 shows an order image between scenario elements in the scenario S2 as a schematic image.

  However, in FIG. 12, the description of the logic of the condition / inference regarding the scenario or scenario element defined in the scenario condition definition information 145 is omitted. In addition, scenario elements are duplicated between scenarios, there is an inclusive relationship between scenarios, and scenarios are dynamically changed according to the logic defined in the scenario condition definition information 145 and user scenarios or scenario elements. Also good.

  FIG. 13 is a diagram illustrating an example of a schematic image of user scenario information according to the embodiment. From the user scenario history information 147, there are scenario elements arranged in time series as a1 → a2 →... As the action process so far of the user whose user ID is 1001.

  FIG. 13 shows the scenario information of the user whose user ID is 1001, that is, the order relationship of the scenario elements. The area 511 is the entire user scenario information of the user, and the area 512 has been described so far. The portion that can be calculated as a scenario candidate in the usage history (behavior process) is shown as a schematic image.

  However, in FIG. 13, description of information related to time 1472 recorded in the user scenario history information 147 is omitted. Note that the scenario candidate calculation shown in the area 512 will be described with reference to the processing flow of FIG. 14 below. However, each time the user uses the consultation window, that is, every time step 132 in FIG. Dynamic calculation is performed using the information 143, the scenario order definition information 144, and the scenario condition definition information 145.

  At the stage of consultation with the person in charge at the consultation desk, the user's past action history (user scenario information as shown in Fig. 13), and if the persona is linked to the user It has scenario information related to a persona (persona scenario information as shown in FIG. 12) and is managed in the storage function 14 of the server 1.

FIG. 14 is a diagram illustrating a sub-processing flow for calculating user scenario candidates in the server 1 of the embodiment.
FIG. 14 is a processing flow for calculating a scenario candidate corresponding to the user while performing a question to the user in step 132 of FIG.

  In step 1321, the scenario configuration information 143, the scenario order definition information 144, and the scenario condition definition information 145 are referred to the user scenario history information 147 of the user, and a scenario is selected from the scenario information of the user. Calculate whether candidates can be extracted.

  If a scenario candidate can be extracted in this step, the scenario candidate information is retained in the subsequent processing. Note that there are a plurality of scenario candidates, and these may be retained, or some of the candidates may be selectively retained. In this case, when the calculated value is calculated by the above calculation, the value may be selected based on the magnitude of the value (for example, a predetermined threshold value or more and a certain number or more). An area 512 in FIG. 13 is an example showing that the scenario S0 can be extracted as a scenario candidate from the scenario information of the user.

  In step 1322, one scenario is selected from the scenario candidates extracted in step 1321, and the user can be executed with reference to the user scenario history information 147 and question definition information 148 of the user. A set of questions. As a method of selecting one scenario from the scenario candidates, for example, referring to the weight 1435 of the scenario configuration definition information 143, the sum of the weights of the scenario elements existing in the scenario information of the user is the largest. A method of selecting a certain scenario can be considered.

  For the selected scenario, the question definition information 148 is referenced to extract a set of questions that have not been asked to the user. If the answer is already known without executing the question from the user scenario information, such a question is excluded from the set of questions. For example, if a scenario element “Investment trust sale check” already exists in the scenario information of the user, even if the question definition information 148 has a question “Do you consider selling an investment trust?” Excluded from the set.

  On the other hand, if there is no scenario candidate in step 1321, the question definition information 148 is referred to, a question that has not yet been asked to the user from a set of all questions related to the persona, and a scenario of the user that has already been made. Excludes questions whose answers are known without having to ask questions from the information to make a set of questions. However, since it is necessary to satisfy the question conditions specified in the condition 1485 of the question definition information 148, questions that do not satisfy the condition are excluded from the set of questions. For example, if you are already asking questions for users who are already in the scenario of “mortgage contracts by sale of investment trusts”, ask questions related to other scenarios, or even the same scenario at the initial stage (faster in the order of questions) Questions that date back to the stage are excluded.

  In step 1323, if there is a set of questions extracted in step 1322, the process proceeds to step 1324, and if there is no set of questions (if empty), all questions that can be performed by the user are finished. The process shown in FIG. 14 is completed, and the process proceeds to step 133 in FIG.

In step 1324, one optimal question to be executed next is extracted from the set of questions extracted in step 1322, and the user is asked a question.
As a method for extracting one question, for example, a method of referring to the condition 1485 of the question definition information 148 and extracting a question to be performed first in the current set of questions as a question order relationship is considered. It is done. This is determined according to information indicating the priority included in the condition 1485 of the question definition information 148.

  In the case where there is no scenario candidate in the step 1321, a method of preferentially extracting the question when the question common to each scenario exists with reference to the question definition information 148 can be considered.

FIG. 15 is a diagram illustrating an example of a hearing support screen for a user in the window terminal 3 according to the embodiment.
FIG. 15 is a screen for displaying the question information extracted in step 132 to the person in charge of the consultation desk and supporting the hearing to the user.
When a persona is associated with the user, the screen area 611 displays information about the persona associated with the user (for example, an outline of the persona).

The screen area 612 displays a question ID, a question content, an answer option, and the like as information on a question to be conducted to the user. Depending on the content of the user's answer, the person in charge of the consultation desk inputs an answer using the input function 31 of the window terminal 3. Alternatively, a terminal that can be input by the user is installed separately, and the answer column in FIG. 15 is input in conjunction with input to the terminal (selection of answer to a question). Also good.
The screen area 613 displays guidance (advice) that supports customer service for a consultant.

  In the exchange of questions between the user and the consultation service representative, a screen as shown in FIG. 15 is displayed to the consultation service representative whenever the next question is extracted according to the user's answer.

  In step 1325, a scenario element indicating a user's behavior and needs is newly added by referring to the semantic content of the scenario element described in the scenario element definition information 142 by answering the question in step 1324. If this is possible, the scenario element is retained as a scenario element that can be used for calculation of a scenario candidate of the user.

  In step 1326, as in step 1321, the user scenario history information 147 of the user is referred to the scenario configuration definition information 143, scenario order definition information 144, and scenario condition definition information 145, and the user It is calculated whether or not a scenario candidate can be extracted from the scenario information (determined that it can be extracted if the calculation result matches a predetermined condition such as within a predetermined range). In this calculation, if there is a scenario element that can be newly utilized in the step 1325, it is calculated whether or not a scenario candidate can be extracted by combining this scenario element in addition to the user scenario history information 147 of the user. To do.

FIG. 16 is a diagram illustrating an example of a schematic image of a user scenario candidate according to the embodiment.
When a scenario element a6 (522 in FIG. 16) can be newly added to the scenario information of the user in the step 1325, the scenario S1 which is a user scenario candidate is as shown by 521 in FIG. Alternatively, as another case, a scenario S1 can be newly extracted as a scenario candidate by adding the scenario element a6 by the condition / inference 1453 of the scenario condition definition information 145.

  It is assumed that a new scenario element b2 (524 in FIG. 16) can be added by adding the scenario element a6 as shown in FIG. 16 and referring to the condition / inference 1453 of the scenario condition definition information 145. Furthermore, it is assumed that scenario S2 (523 in FIG. 16) can be extracted as a scenario candidate by adding the scenario element b2 and referring to the condition / inference 1453 of the scenario condition definition information 145. In this inference logic, it is possible to calculate which scenario has priority by calculating the sum of the weights of the scenario elements existing in each scenario with reference to the weight 1435 of the scenario configuration definition information 143. . For example, only scenario S1 has been extracted as a scenario candidate, but scenario S2 is newly extracted as a scenario candidate by the above example, the sum of weights of scenario S1 is 70, and the sum of weights of scenario S2 is 110. In some cases, it can be predicted that the newly extracted scenario S2 has a higher priority, that is, a scenario closer to the potential needs of the user.

  With the inference logic as described above, not only S1 but also S2 can be newly extracted as a scenario candidate for the user, and it is possible to find out the potential needs of the user.

  When the process in this step is completed, the process returns to step 1322 and the process is repeated. In addition, when the user's scenario candidate can be narrowed down to one by the process in this step, the process flow in FIG. 14 may be interrupted and the process may proceed to step 133 in FIG. When the sub-processing flow for calculating the user scenario candidate in FIG. 14 is completed, the process proceeds to step 133 in FIG.

  In step 133, the calculation result of the scenario candidate extracted in FIG. 14 is referred to, and it is checked whether or not there is one or more scenario candidates for the user. If there is no scenario candidate, the process proceeds to step 135.

  In step 134, a plurality of scenario candidates are shown to the user, a scenario selection is requested, and one scenario is determined. Alternatively, as described in step 1326 of FIG. 14, the sum of the weights of each scenario candidate is calculated, and the scenario predicted to be closer to the potential needs of the user is presented to the user and predicted. It may be a method to check whether the scenario is correct or not. That is, the selection is made according to the input through the user's input device.

FIG. 17 is a diagram illustrating an example of a proposal support screen for a user in the window terminal 3 according to the embodiment. FIG. 17 is a screen for displaying a specific proposal for the user to the person in charge of the consultation desk based on the scenario determined in this step.
When a persona is associated with the user, the screen area 621 displays information about the persona associated with the user (for example, an outline of the persona).

  The screen area 622 displays information related to the proposal to the user. Specific proposal contents shall be prepared in advance for each scenario of each persona. The screen area 623 displays guidance (advice) for supporting the customer service for the person in charge of the consultation service.

  In step 135, when there is no scenario candidate, in order to calculate the scenario candidate, it is necessary to individually interview the user and enrich the scenario information.

  In step 136, when no persona is associated with the user, scenario information of an arbitrary persona, scenario information about another user, anonymous scenario information that cannot identify the user is used as alternative scenario information, and Calculate user scenario candidates.

  Next, anonymous scenario information in which a user cannot be identified will be described. When a specific user can be specified by the user ID, the usage history can be acquired like the usage history information 141. On the other hand, when the user refers to the WWW page without going through personal authentication, information such as the IP address of the user terminal 2 used by the user is obtained instead of the user ID 1412 of the usage history information 141. can do. Such information such as an IP address can be acquired by a function of a general WWW server. If there is access by the same IP address, a specific user cannot be specified, but it is determined that there is a high possibility that the same user is accessing, and scenario information obtained from such a user is obtained. This is referred to as anonymous scenario information (for such anonymous scenario information, information (flag or the like) indicating it is associated).

  In this step, the behavior process closer to the user is considered by using the scenario information of any persona, scenario information about other users, and anonymous scenario information in the same manner as the process of step 132 (process of FIG. 14). Candidate scenarios can be calculated with reference to scenario information of people with

  As described above, in this embodiment, a user's action type is modeled using a plurality of scenarios. Specifically, the relationship between scenarios and the relation between specific actions within a scenario are expressed by logic ( Program), and the user's specific scenario candidates are calculated based on the behavior type to which the modeled user belongs, the user's past behavior history (behavior process), and questions to the user. It is possible to capture potential usage needs that take into account the essential background of user behavior, not usage needs derived from superficial behavior of users.

  It can be used as a device, method, and program that supports consulting services at the consultation counter of a bank sales office, for example, where communication occurs between users and salesclerks or representatives in the fields of finance, public, distribution, etc. It is.

1 server 2 user terminal 3 window terminal 4 communication network

Claims (12)

In an information processing device that can provide information according to the user's latent needs corresponding to the user's behavior type and behavior process,
A persona that is information indicating the user's behavior type and behavior process, scenario configuration definition information that defines a scenario component indicating a usage scene in a certain aspect of the user, and an occurrence order of the scenario components Means for storing scenario order definition information for defining relevance;
Means for inputting a response to the question to the user;
Means for generating a scenario for the user using the input response and the stored content in the means for storing;
The information processing apparatus characterized in that the generating means changes the generated scenario based on a response further input by the inputting means and further creates a new scenario. The information processing apparatus according to claim 1,
An information processing apparatus further comprising means for executing a question to the user when the generating means generates the new scenario. The information processing apparatus according to claim 1,
Means for presenting the generated scenario to the user if the generating means fails to generate the new scenario;
An information processing apparatus further comprising means for specifying the scenario as a final scenario based on an instruction from the user with respect to the presented scenario. The information processing apparatus according to claim 1,
The means for generating is
It is determined whether a plurality of scenario candidates can be calculated according to the means for storing,
If the scenario candidate can be calculated, calculate each scenario candidate, identify a question to the user for each calculated scenario candidate,
Identify the scenario components that correspond to the responses to the identified questions,
An information processing apparatus that generates the new scenario using the identified component of the scenario. The information processing apparatus according to claim 1,
The information generating apparatus, wherein the generating unit generates the scenario using anonymous information that cannot be specified by the user when the persona is not stored. The information processing apparatus according to any one of claims 1 to 5,
The means for storing further stores scenario condition definition information that defines a condition to be satisfied for the persona or an inference condition describing a result obtained by inference and a constraint condition related to the scenario as formal logic,
The information generating apparatus is characterized in that the generating means generates the scenario by predicting the needs or actions of the user using the scenario condition definition information. An information processing device that can provide information according to the user's potential needs corresponding to the user's behavior type and behavior process,
A persona that is information indicating the user's behavior type and behavior process, scenario configuration definition information that defines a scenario component indicating a usage scene in a certain aspect of the user, and an occurrence order of the scenario components Means for storing scenario order definition information for defining relevance;
Means for inputting a response to the question to the user;
Using the input response and the content stored in the storage means, function as means for generating a scenario for the user,
A program that realizes a function of generating a new scenario by changing the generated scenario based on a response further input by the input unit. The program according to claim 7,
A program for further causing a function to execute a question to the user when the generating unit generates the new scenario. In the program according to claim 7 or 8,
Means for presenting the generated scenario to the user if the generating means fails to generate the new scenario;
A program for further causing a means for specifying the scenario as a final scenario according to an instruction from the user with respect to the presented scenario. The program according to any one of claims 7 to 9,
In the means for generating,
It is determined whether a plurality of scenario candidates can be calculated according to the means for storing,
If the scenario candidate can be calculated, calculate each scenario candidate, identify a question to the user for each calculated scenario candidate,
Identify the scenario components that correspond to the responses to the identified questions,
A program for causing a function to generate the new scenario to be executed using the identified component of the scenario. The program according to any one of claims 7 to 10,
A program for causing the generating means to execute a function of generating the scenario using anonymous information that cannot be specified by the user when the persona is not stored. The program according to any one of claims 7 to 11,
The storage means further stores scenario condition definition information that defines a condition to be satisfied for the persona or an inference condition describing a result obtained by inference and a constraint condition related to the scenario as formal logic,
A program that causes the generating means to generate the scenario by predicting the needs or actions of the user using the scenario condition definition information.

Images