Proposal device, proposal method, and program

Abstract

The objective of the present disclosure is to facilitate acceptance by a user of a proposal content related to behavior modification.　Accordingly, the present disclosure is a proposal device that makes a proposal for behavior modification to a user in dialog format by an agent, the proposal device comprising: a behavior modification assistance stage management unit that manages behavior modification assistance stage information indicating a behavior modification assistance stage for the user, on the basis of the degree of trust the user has with respect to the agent; a calculation unit that calculates a self-disclosure score that is a value corresponding to the number of disclosures a predetermined user has made, on the basis of self-disclosure data related to the predetermined user; a determination unit that determines a behavior modification assistance stage as a transition destination corresponding to the self-disclosure score from the behavior modification assistance stage information; a selection unit that selects a specific proposal content according to the behavior modification assistance stage as the transition destination from among a plurality of proposal contents corresponding to the behavior modification assistance stage; and a transmission unit that transmits data of the specific proposal content to a predetermined user terminal of the predetermined user.

Description

Proposed device, proposed method, and program

[0001] The present disclosure relates to a technology for providing a user with suggestions for behavioral modification regarding a predetermined behavior.

[0002] In modern society, it is important to support users in changing their own behaviors, such as preventing lifestyle-related diseases. For this reason, there are systems (AI-based agents) that provide health advice (Non-Patent Document 1). In addition, conventional systems provide users with suggestions to support behavioral change by following a predetermined scenario (Non-Patent Document 2).

[0003] "Behavioral change methods for improving QOL using artificial intelligence and life logs" (2017)<https: / / www.jstage.jst.go.jp / article / jahpp / 30 / 0 / 30_126 / _pdf / -char / ja> Koyo Otsu, Yuki Nishida, Keita Kiuchi, Yugo Hayashi, Introduction of conversational tasks for realizing personalized healthcare using chatbots (2022)

[0004] However, when the user has low trust in the agent or low intimacy with the agent, even if the suggestion is made to the user according to a predetermined scenario, the impact of the suggestion is small, and it often does not lead to a change in the user's behavior.

[0005] The present disclosure aims to solve the above-mentioned problems by making it easier for users to accept suggestions regarding behavioral change.

[0006] In order to achieve the above-mentioned object, the present disclosure provides a proposal device that makes suggestions for behavior change to a user in an interactive format using an agent, the proposal device having: a behavior change support stage management unit that manages behavior change support stage information that indicates the behavior change support stage for the user based on the trust the user has in the agent; a calculation unit that calculates a self-disclosure score, which is a value corresponding to the number of times a specified user has disclosed, based on self-disclosure data about the specified user; a determination unit that determines a destination behavior change support stage based on the self-disclosure score from the behavior change support stage information; a selection unit that selects a specific proposal content from a plurality of proposal contents corresponding to the behavior change support stage, based on the destination behavior change support stage; and a transmission unit that transmits data of the specific proposal content to a specified user terminal of the specified user.

[0007] As described above, the present disclosure has the effect of making it easier for users to accept suggestions regarding behavioral modification.

[0008] 1 is an overall configuration diagram of a communication system according to an embodiment. FIG. 1 is an electrical hardware configuration diagram of a mobile terminal, an operation terminal, a dialogue device, and a database server according to an embodiment. FIG. 2 is a functional configuration diagram of a communication system according to an embodiment. FIG. 3 is a diagram showing explicit self-disclosure data. FIG. 4 is a diagram showing optical fiber information managed in a fiber information management unit. FIG. 5 is a diagram showing user information (part 1) managed in a user information DB. FIG. 6 is a diagram showing user information (part 2) managed in a user information DB. FIG. 7 is a diagram showing user information (part 3) managed in a user information DB. (a) is a conceptual diagram showing the concept of tf-idf, and (b) is a conceptual diagram showing application and arrangement to this embodiment. FIG. 8 is a diagram showing behavior change support stage management information managed in a behavior change support stage DB. FIG. 9 is a diagram showing a behavior change support stage (state) map managed in a proposed scenario DB. FIG. 10 is a diagram showing a transition condition management table managed in a proposed scenario DB. FIG. 11 is a diagram showing an example of a behavior tree in which proposed contents for state S1 are defined. FIG. 12 is a diagram showing an example of a behavior tree in which proposed contents for state S2 are defined. FIG. 13 is a diagram showing an example of a behavior tree in which proposed contents for state S3 are defined. FIG. 1 is a diagram showing an example of a behavior tree in which proposal contents related to state S4 are defined. FIG. 2 is a diagram showing an example of a behavior tree in which proposal contents related to state S5 are defined. FIG. 3 is a diagram showing an example of a behavior tree in which proposal contents related to state S6 are defined. FIG. 4 is a diagram showing an example of a behavior tree in which proposal contents related to state S7 are defined. FIG. 5 is a diagram showing an example of a behavior tree in which proposal contents related to state S8 are defined. FIG. 6 is a diagram showing an example of a screen of an operation terminal. FIG. 7 is a flowchart showing processing related to an embodiment.

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the present invention is not limited to the embodiments shown below, and various modifications are possible within the scope of the technical concept of the present invention. Since the drawings are intended to conceptually explain the present invention, dimensions, ratios, or numbers may be exaggerated or simplified as necessary to facilitate understanding.

[0010] [System Configuration of the Embodiment] First, the overall configuration of a communication system according to the embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing the overall configuration of a communication system according to the embodiment.

[0011] 1 , the communication system 10 of this embodiment is constructed by a mobile terminal 20, an operation terminal 30, an interaction device 50, and a database server 70. The mobile terminal 20 and the operation terminal 30 are owned and operated by a predetermined user Y. The mobile terminal 20 and the operation terminal 30 are examples of user terminals, and the mobile terminal 20 may also function as the operation terminal 30.

[0012] The dialogue device 50 is composed of one or more computers. When the dialogue device 50 is composed of multiple computers, it may be referred to as a "dialogue device" or a "dialogue system." The dialogue device 50 provides a dialogue service that makes suggestions for behavioral modification related to a predetermined behavior to the user through a dialogue format using an agent such as AI (Artificial Intelligence) possessed by the dialogue device 50. Specifically, the dialogue device 50 transmits data on the content of suggestions to the mobile terminal 20 and the operation terminal 30 to support behavioral modification related to a predetermined behavior that the user should work on, such as prevention of lifestyle-related diseases. In this case, in order to make it easier for the user to accept the content of suggestions related to behavioral modification, the dialogue device 50 performs the following processing.

[0013] The dialogue device 50 calculates a "self-disclosure score" that represents a value corresponding to the number of times that user Y has disclosed information about at least one of the four categories of user Y's lifestyle habits and predetermined conditions (physical condition, psychological condition, and social condition) using self-disclosure data, i.e., a value that represents the degree of user Y's openness to the dialogue device 50 (agent).

[0014] Next, the dialogue device 50 uses the "self-disclosure score" to quantify the trust (degree of intimacy) that user Y has with the agent, and thereby determines the "behavioral change support stage," which is the stage of behavioral change support to be proposed to user Y. Next, the dialogue device 50 selects proposal content according to the "behavioral change support stage," and proposes it to user Y.

[0015] Here, the reason for using the "self-disclosure score" and "behavioral change support stage" in this embodiment will be explained. In the person-to-person motivational dialogue implemented under Japan's specific health checkup and specific health guidance system, building trust is said to be important, and this is incorporated into motivational interviews conducted by professionals (see Reference 1). Furthermore, the interviewer determines whether a relationship has been established with the subject based on the subject's comments and facial expressions, and the dialogue proceeds according to the depth of the relationship (see Reference 2). Health guidance implemented based on this specific health guidance system has been shown to be somewhat effective (see References 3 and 4). This embodiment applies this concept to suggestions regarding user behavioral change. (Reference 1) Ministry of Health, Labour and Welfare, Standard Health Checkup and Health Guidance Program [2018 Edition], pp.3-56-3-61. (Reference 2) Sato et al., "Analysis of Motivational Dialogue Process in Health Guidance for the Prevention of Lifestyle-Related Diseases." (Reference 3) Tsushita, K., Hosler, A., Miura, K., Rationale and Descriptive Analysis of Specific Health Guidance: the Nationwide Lifestyle Intervention Program Targeting Metabolic Syndrome in Japan. Observational Study J Atheroscler Thromb; Vol. 25, No. 4, pp.3088-322 (2018). (Reference 4) Tsugawa, Y., Fukuma, S., Latest Evidence on the Effectiveness of Specific Health Checkups and Specific Health Guidance, Shinzou, Vol. 53, No. 5, pp. 417-423 (2021). The dialogue device 50 may unilaterally make suggestions for behavioral modification to the user, rather than in a dialogue format. The dialogue device 50 is an example of a suggestion device.

[0016] The database server 70 is a server that manages a DB (Data Base) required for the dialogue device 50 to select proposal contents.

[0017] The interactive device 50 is capable of data communication with the mobile terminal 20, the operation terminal 30, and the database server 70 via a communication network 100 such as the Internet. The communication may be wired or wireless.

[0018] [Hardware Configuration] Next, the hardware configuration of the mobile terminal 20 will be described with reference to Fig. 2. Fig. 2 is a diagram showing the hardware configuration of the mobile terminal according to the embodiment.

[0019] 2, the mobile terminal 20 has a processor 1001, a memory 1002, an auxiliary storage device 1003, a communication device 1004, and a connection device 1005. The mobile terminal 20 also has an audio input device 1006, an audio output device 1007, a display device 1008, and an imaging device 1009. The hardware components constituting the mobile terminal 20 are connected to each other via a bus 1010 such as a data bus.

[0020] The processor 1001 serves as a control unit that controls the entire mobile terminal 20, and includes various computing devices such as a CPU (Central Processing Unit). The processor 1001 reads and executes various programs on the memory 1002. The processor 1001 may also include a GPU (Graphics Processing Unit).

[0021] The memory 1002 has a main storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The processor 1001 and the memory 1002 form a so-called computer, and the processor 1001 executes various programs read onto the memory 1002, thereby enabling the computer to realize various functions.

[0022] The auxiliary storage device 1003 stores various programs and various information used when the various programs are executed by the processor 1001 .

[0023] The communication device 1004 is a communication device for transmitting and receiving various types of information to and from other devices (including equipment, servers, and systems).

[0024] The connection device 1005 is a connection device used when connecting various sensors, external memories, etc. to the mobile terminal 20 .

[0025] The audio input device 1006 detects audio information such as the user's voice, surrounding sounds, etc. The audio output device 1007 is a device that outputs, by audio, various types of information received from other devices, for example.

[0026] The display device 1008 is, for example, a device (such as a display) that displays images of various information received from other devices.

[0027] The imaging device 1009 captures images of the user and the surroundings and generates image information.

[0028] The operation terminal 30, the dialogue device 50, and the database server 70 have the same configuration as those in Fig. 2, and therefore description thereof will be omitted. The dialogue device 50 and the database server 70 do not necessarily have to include the voice input device 1006 and the voice output device 1007.

[0029] Next, a functional configuration diagram of the communication system according to the embodiment will be described with reference to Fig. 3. Fig. 3 is a functional configuration diagram of the communication system according to the embodiment.

[0030] [Functional Configuration of Mobile Terminal] As shown in Fig. 3, the mobile terminal 20 has a data transmission unit 21 and a detection unit 22. These units each have a function that is realized by instructions from the processor 1001 in Fig. 2 based on a program.

[0031] The data transmission unit 21 transmits data such as implicit self-disclosure data, which will be described later, to the dialogue device 50 .

[0032] The detection unit 22 detects implicit self-disclosure content related to the user's exercise, sleep, etc., and outputs the detection result, that is, implicit self-disclosure data, to the data transmission unit 21.

[0033] 3, the operation terminal 30 has a data transmission unit 31 and an output unit 32. These units each have a function realized by an instruction from the processor 1001 in FIG. 2 based on a program.

[0034] The data transmission unit 31 transmits explicit self-disclosure data, which will be described later, to the dialogue device 50. This explicit self-disclosure data also includes answers to questions in the proposal content from the dialogue device 50.

[0035] The output unit 32 can display the proposed content (question) and answer as shown in FIG. 12, and can input and output voice instead of or in addition to the display.

[0036] [Functional Configuration of the Interactive Device, DBs in the Database Server] Next, the functional configuration of the interactive device 50 and each DB (Data Base) in the database server 70 will be described with reference to FIG.

[0037] The dialogue device 50 has a data acquisition unit 51, a self-disclosure score calculation unit 52, a behavioral change support stage determination unit 53, a proposed content selection unit 54, and a proposed content transmission unit 55. Each of these units is a function realized by an instruction from the processor 1001 in Fig. 2 based on a program. Note that the data acquisition unit 51, the self-disclosure score calculation unit 52, the behavioral change support stage determination unit 53, the proposed content selection unit 54, and the proposed content transmission unit 55 are examples of an acquisition unit, a calculation unit, a determination unit, a selection unit, and a transmission unit, respectively.

[0038] Furthermore, database server 70 has a self-disclosure data DB 71, a user information DB 72, a self-disclosure score DB 73, a behavioral change support stage DB 74, and a proposed scenario DB 75. Each of these DBs is stored in a storage unit constructed by memory 1002 or auxiliary storage device 1003 in Fig. 2. Note that at least one of the DBs may be stored in dialogue device 50, rather than in database server 70.

[0039] <Data Acquisition Unit> The data acquisition unit 51 acquires user Y's self-disclosure data (FIGS. 4 and 5) from the mobile terminal 20 and stores it in the self-disclosure data DB 71. The self-disclosure data includes explicit self-disclosure data as shown in FIG. 4 and implicit self-disclosure data as shown in FIG. 5. Note that explicit self-disclosure data is necessary as self-disclosure data, but implicit self-disclosure data does not have to be used as self-disclosure data. The acquisition timing is at intervals (such as once a day) that are preset by the dialogue device 50 setting an initial setting value or by user Y setting it when starting to use the dialogue service, or when user Y actively transmits self-disclosure data.

[0040] Furthermore, when the data acquisition unit 51 acquires even one piece of self-disclosure data, it requests the self-disclosure score calculation unit 52 to calculate the self-disclosure score of the user Y.

[0041] (Explicit Self-Disclosure Data) Explicit self-disclosure data is data that user Y actively answers (fills in) and sends during an interactive exchange with an agent. As shown in FIG. 4, explicit self-disclosure data is data relating to at least one of four categories (user Y's lifestyle habits, user Y's specified state (psychological state, physical state, and social state)) that should be considered when encouraging user Y to change their behavior. However, it is not necessary to distinguish between categories. In other words, there may be no category (only one category), or there may be two, three, five or more categories.

[0042] (Implicit Self-Disclosure Data) Implicit self-disclosure data is data that is not actively transmitted by user Y each time, but is automatically transmitted from the mobile terminal 20 used by user Y. It is assumed that explicit permission for data transmission is obtained from user Y when the use of the dialogue service begins, and that explicit transmission is no longer necessary after the permission is obtained. As shown in FIG. 5 , the content of the implicit self-disclosure data is data acquired from the detection unit 22 of the mobile terminal 20 used by user Y or an application installed on the mobile terminal 20. Data from the detection unit 22 includes, for example, step count data measured using an acceleration sensor. Data from an application includes, for example, weight data registered in the application and blood pressure data acquired by the application via a smartwatch or the like. Data also includes data indicating the frequency of interactions with an agent and data indicating user Y's health checkup results registered in the application by user Y. Like explicit self-disclosure data, implicit self-disclosure data includes data related to four categories, but the four categories do not necessarily need to be distinguished.

[0043] <Self-disclosure score calculation unit> (Creation of user information) When self-disclosure score calculation unit 52 receives a calculation request from data acquisition unit 51, it reads out the self-disclosure data of user Y from self-disclosure data DB 71, creates user information ( FIG. 6A , FIG. 6B or FIG. 6C ) by quantifying each item of this self-disclosure data, and overwrites it in user information DB 72. Here, multiple types of user information will be described.

[0044] ((User Information (Part 1))) FIG. 6A is a diagram showing user information (Part 1) managed in the user information DB. For example, as shown in FIG. 6A, the user information shows the user ID, the number of questions from the dialogue service agent, and the number of answers by the user, all associated with each other. Regarding the method of counting the number of answers, for example, in FIG. 12, when any of the options "Western food," "Japanese food," or "I don't eat," is selected, the count is incremented by "1," but when the option "I don't want to answer" is selected, the count is not incremented by "1." This also applies to (Parts 2) and (Part 3) described below.

[0045] ((User Information (Part 2))) FIG. 6B is a diagram showing user information (part 2) managed in the user information DB. As another pattern, as shown in FIG. 6B, the user information may, for example, associate and display the user ID, the number of questions from the dialogue service agent, the number of answers from the user, and the range of answers (number of categories). Note that even if a user answers multiple questions related to the same category, for example, "lifestyle habits," out of the four categories, the number of categories for "lifestyle habits" is "1." On the other hand, if a user answers one question each related to the categories of "lifestyle habits" and "psychological state," the number of categories for "lifestyle habits" and "psychological state" is "2."

[0046] ((User Information (Part 3))) Figure 6C is a diagram showing user information (Part 3) managed in the user information DB. As another pattern, the user information shows, for example, the user ID and the number of questions and the number of answers for each category, in association with each other, as shown in Figure 6C. In the case of (Part 3), unlike (Part 2), if a user answers multiple questions related to the same category, for example, "lifestyle habits," the number of answers will be "multiple."

[0047] (Creating a self-disclosure score) The self-disclosure score calculation unit 52 further calculates a self-disclosure score from the user information using the calculation method shown below. The calculation method for the self-disclosure score differs depending on the type of user information used, as shown below.

[0048] ((First Calculation Method)) The self-disclosure score calculation unit 52 calculates the self-disclosure score based on the "number of responses" of the user information (1) or the user information (2), as shown below.

[0049] Score j = User i For example, in FIGS. 6A and 6B, the self-disclosure score of the user with user ID "001" is "3."

[0050] ((Second Calculation Method)) As shown below, the self-disclosure score calculation unit 52 calculates a self-disclosure score of a predetermined user from the frequency of self-disclosure relative to the number of questions (response rate) by comparing the frequency of disclosure (number of answers) with the number of times other users have disclosed based on the "number of questions" and "number of answers" in user information (part 1) or user information (part 2). That is, the self-disclosure score calculation unit 52 calculates a self-disclosure score that is a value corresponding to the disclosure rate (response rate) disclosed by the predetermined user relative to the disclosure rate (response rate) disclosed by all users using self-disclosure data for at least one category. For example, even if the response rate of the predetermined user is low, if the response rates of other users are also low, the self-disclosure score of the predetermined user will not be relatively low.

[0051] In this case, the self-disclosure score calculation unit 52 uses the standard deviation of the response rate as shown below.

[0052]

[0053] For example, in FIGS. 6A and 6B, the response rate of each user is calculated as follows:

[0054] Response rate of user with user ID "001" = 0.6 Response rate of user with user ID "002" = 1.0 Response rate of user with user ID "003" = 0.8 This results in the following values ​​being calculated.

[0055] Therefore, the self-disclosure score for user ID "001" is Score1 = (0.6 - 0.8) / 0.16 = -1.22.

[0056] In addition, in the first or second calculation method, the self-disclosure score calculation unit 52 may calculate the self-disclosure score of a specified user based on the “number of categories” instead of the “number of responses” in the user information (part 2).

[0057] ((Third Calculation Method)) The self-disclosure score calculation unit 52 calculates the self-disclosure score for each category based on the user information (part 3), as shown below. In this case, the self-disclosure score calculation unit 52 derives the category of words included in each question content using the formulas shown below, and calculates the self-disclosure score of each user for each category using the formulas shown below.

[0058]

[0059]

[0060] category i The self-disclosure score of i, j is Score(i, j) = tf·idf·ar. Here, tf-idf when deriving the category of words included in each question content will be explained using Fig. 7. Fig. 7 is a conceptual diagram showing the idea behind tf-idf (a) and its application and arrangement to this embodiment (b).

[0061] The concept of tf-idf is expressed by tf * idf in Figure 7(a). tf: Each document D j Words in i IDF: How often a word appears in all documents i tf: User Y j In the category i IDF: How many times the answer about a category appears when looking at all users i However, the tf-idf applied in this embodiment does not have a feature regarding the number of answers to the number of questions. jHowever, it does not indicate whether the user is the type who answers questions often or the type who does not answer (does not want to answer). Therefore, as an adjustment when applying this, the above calculation formula is added with the consideration of the number of answers relative to the number of questions. Note that in the above calculation formula, the "Y" in "User Y" is omitted.

[0062] For example, in FIG. 12 , in the case of the question "What do you usually eat for breakfast?", the word "breakfast" is determined to be a word included in the category "lifestyle." If user Y selects the option "Western food," "Japanese food," or "don't eat," the number of selections in the category "lifestyle" is incremented by one. However, even if user Y selects the option "don't want to answer," the number of selections in the category "lifestyle" is not incremented. Note that if a question contains a word included in two or more categories, when an option other than "don't want to answer" is selected, the number of selections for each of the two or more categories is incremented by one.

[0063] <Behavioral change support stage determination unit> (Determining the level of the self-disclosure score) When a determination request is received from the self-disclosure score calculation unit 52, the behavioral change support stage determination unit 53 reads the self-disclosure score corresponding to the user ID of user Y from the self-disclosure score DB 73, and determines the level of the self-disclosure score from the self-disclosure score using the level determination conditions shown below. The "level of the self-disclosure score" is a value indicating the category that includes the self-disclosure score out of a predetermined number of categories indicated in stages.

[0064] ((Level determination condition (1))) Lv1: Self-disclosure score < 1 Lv2: 1 ≦ self-disclosure score < 3 Lv3: 3 ≦ self-disclosure score ((Level determination condition (2))) In addition, the average and standard deviation of the self-disclosure scores of multiple users may also be used, as shown below.

[0065] Lv1: Self-disclosure score < [average] - [standard deviation] Lv2: [average] - [standard deviation] ≦ Self-disclosure score < [average] + [standard deviation] Lv3: [average] + [standard deviation] < Self-disclosure score (Reading the current behavior change support stage) The behavior change support stage determination unit 53 reads information indicating the current behavior change support stage (state) corresponding to the user ID of user Y from the behavior change support stage DB 74. The initial value of the behavior change support stage is "S1".

[0066] (Behavior support change stage DB) Figure 8 is a diagram showing behavior support change stage management information managed in the behavior support change stage DB. As shown in Figure 8, the behavior support change stage management information shows a user ID and information indicating the current behavior change support stage (state) in association with each other.

[0067] (Determination of behavioral change support stage) Next, the behavioral change support stage determination unit 53 determines a behavioral change support stage to transition from the current behavioral change stage to based on the level of the self-disclosure score, assuming that the proposal for the current behavioral change support stage has been completed, using the transition condition management table (see Figure 10) read from the proposed scenario DB 75. Note that if the self-disclosure score is expressed as one of multiple integer values, the behavioral change support stage determination unit 53 may determine the behavioral change support stage to transition to by using the self-disclosure score as is, rather than the level of the self-disclosure score.

[0068] Here, the proposed scenario DB 75 will be explained using Figures 9 to 11 (Figures 11A to 11H). The proposed scenario DB 75 is broadly composed of three types of data. The first is the behavior change support stage map shown in Figure 9, the second is the transition condition management table shown in Figure 10, and the third is the behavior tree shown in Figures 11A to 11H. These three types of data will be explained in detail below.

[0069] (Behavioral Change Support Stage Map) As shown in Figure 9, the behavioral change support stage map shows the relationship between the current behavioral change support stage and the next behavioral change support stage, and is a state transition diagram that shows the overall picture of the story of how the dialogue between user Y and the agent progresses.

[0070] (Transition Condition Management Table) The transition condition management table indicates the transition conditions at each behavior change support stage in the behavior change support stage map and the transition destination when the transition conditions are satisfied.

[0071] The "transition conditions" in the transition condition management table are basically set so that the behavioral change support stage to which the user transitions increases as the self-disclosure score increases. However, for example, if the current behavioral change support stage is "S5," the behavioral change support stage may be lowered to "S2" or "S3" depending on the transition conditions. This is not because the relationship between user Y and the agent has deteriorated, but because the time spent at behavioral change support stage "S5" has exceeded the required time, and the user has stayed at behavioral change support stage "S5" for too long, causing the user to be moved to another behavioral change support stage.

[0072] (Behavior Tree) A behavior tree is data showing suggested dialogue content for behavior change support stages "S1" to "S8" in the behavior change support stage map. The suggested dialogue content for behavior change support stages "S1" to "S8" is shown in Figures 11A to 11H, respectively, and is selected by the suggested content selection unit 54.

[0073] Here, in FIGS. 11A and 11H, the proposed content (for example, a message text) is determined, and each proposed content is selected.

[0074] In FIG. 11B , "*" indicates priority, with groups with a larger number of "*"s being selected first. For example, after the selection of predetermined suggested content (questions and options) from the psychological state group with the highest priority is completed in "S2," other suggested content is selected from the lifestyle habit group with the next highest priority in the next "S2." Alternatively, after the proposal of all suggested content from the psychological state group with the highest priority is completed in "S2," all suggested content is selected from the lifestyle habit group with the next highest priority in the next "S2." Furthermore, each group is composed of multiple categories, such as a diet category and an exercise category, and each category is selected randomly. Furthermore, each category is composed of multiple suggested content (questions and options), and each suggested content is also selected randomly.

[0075] Each suggestion (question and option) shown in FIG. 11C is selected in numerical order.

[0076] Each suggestion (eg, message text) shown in FIG. 11D is randomly selected.

[0077] In FIG. 11E, the proposed content (for example, message text) that meets the following conditions is selected: 5-1 "," "d 5-2 "," "d 5-3 The conditions for each arrow to "d5" are as follows: 5-1 " : IF fasting blood glucose > 100 AND breakfast time = "later than" "d5" → "d 5-2 ": IF fasting blood glucose > 100 AND breakfast content = "Western food" "d5" → "d 5-2 ": IF fasting blood glucose > 100 AND exercise = "less than once a week" Note that "fasting blood glucose," "breakfast time," "breakfast contents," and "exercise" are the contents of the self-disclosure data managed in the self-disclosure data DB71. "Fasting blood glucose" and "exercise" are included in the implicit self-disclosure data shown in Figure 5, but if user Y discloses them himself by operating mobile terminal 20, they will be included in the explicit self-disclosure data shown in Figure 4.

[0078] 11E, further conditions may be added using AND. For example, IF fasting blood glucose > 100 AND breakfast time = "later" AND exercise = "less than once a week." In this case, suggestions can be made to user Y that take into account both diet and exercise, taking into account fasting blood glucose.

[0079] Each suggestion content (for example, a message text) shown in FIGS. 11F and 11G is selected randomly.

[0080] (Updating the behavior change support stage) The behavior change support stage determination unit 53 updates the behavior change support stage to the latest one by overwriting information (behavior change support stage information) indicating the behavior change support stage after the determination (transition destination) in the behavior change support stage DB 74. Note that, as described above, the "behavior change support stage information" is information indicating the behavior change support stage for user Y based on the trust (intimacy) held with user Y agent. In addition, the behavior change support stage determination unit 53 outputs information (behavior change support stage information) indicating the behavior change support stage after the determination (transition destination) to the proposal content selection unit 54.

[0081] <Proposal Content Selection Unit> The proposal content selection unit 54 selects specific proposal content from the behavior tree (Figures 11A to 11H) according to the behavior change support stage after the determination (transition destination), either uniquely (Figures 11A and 11G), randomly taking priority into consideration (Figure 11B), in numerical order (Figure 11C), randomly (Figures 11D and 11F), or to meet specified conditions using self-disclosure data (Figure 11E). The proposal content selection unit 54 then outputs data of the selected specific proposal content to the proposal content transmission unit 55.

[0082] That is, the proposal content selection unit 54 selects a specific proposal content (in) the action support stage information after the determination (transition destination) from among multiple proposal contents corresponding to the action support stage. In this case, in the case of a behavior tree (FIGS. 11A, 11B, 11D, 11F), the proposal content selection unit 54 selects multiple predetermined proposal contents corresponding to the action support stage information after the determination (transition destination) from among multiple proposal contents corresponding to the action support stage, and randomly selects a specific proposal content. Furthermore, when multiple predetermined proposal contents are categorized into multiple categories as in the behavior tree (FIG. 11B), the proposal content selection unit 54 selects a predetermined category according to the priority of the category, and randomly selects a specific proposal content within the predetermined category.

[0083] Furthermore, the suggestion content selection unit 54 may add value to a specific suggestion content, change the expression of a specific suggestion content, or convert a specific suggestion content (text) into voice data by utilizing a system external to the dialogue device 50. For example, in the case of a suggestion content that indicates the intake of foods that are high in iron, the suggestion content selection unit 54 may search for specific foods that are high in iron in an external system and propose them as added value. Furthermore, the suggestion content selection unit 54 may use an external system to change the expression of the suggestion content to female speech, dialect, or the like.

[0084] <Proposal Content Transmission Unit> The proposal content transmission unit 55 transmits data of the specific proposal content selected by the proposal content selection unit 54 to the operation terminal 30. As a result, the output unit 32 of the operation terminal 30 outputs and displays the specific proposal content made by the agent, as shown in Fig. 12. Fig. 12 is a diagram showing an example screen of the display terminal. Note that if the proposal content is audio data, the output unit 32 outputs audio indicating the specific proposal content.

[0085] [Processing of the embodiment] S11: The data acquisition unit 51 acquires the self-disclosure data (FIGS. 4 and 5) of the user Y from the mobile terminal 20 and stores it in the self-disclosure data DB 71.

[0086] S12: When a calculation request is received from the data acquisition unit 51, the self-disclosure score calculation unit 52 reads out user Y's self-disclosure data from the self-disclosure data DB 71, generates user information ( FIG. 6A , FIG. 6B, or FIG. 6C ) in which each item of this self-disclosure data is quantified, and overwrites the user information DB 72. Furthermore, the self-disclosure score calculation unit 52 calculates user Y's self-disclosure score from the user information using the self-disclosure score calculation formula described above. That is, the self-disclosure score calculation unit 52 calculates a self-disclosure score, which is a value corresponding to the number of times user Y has disclosed, using self-disclosure data, at least one of the four categories of the user's lifestyle, physical state, psychological state, and social state. Then, the self-disclosure score calculation unit 52 stores the self-disclosure score in the self-disclosure score DB 73.

[0087] S13: When a determination request is received from the self-disclosure score calculation unit 52, the behavioral change support stage determination unit 53 reads the self-disclosure score corresponding to user Y's user ID from the self-disclosure score DB 73, and determines the current level of the self-disclosure score based on the self-disclosure score using the level determination conditions described above. The behavioral change support stage determination unit 53 also reads behavioral change support stage information indicating the current behavioral change support stage corresponding to user Y's user ID from the behavioral change support stage DB 74. Next, the behavioral change support stage determination unit 53 uses the transition condition management table (see FIG. 10 ) read from the proposed scenario DB 75 to determine a destination behavioral change support stage from the current behavioral change stage based on the level of the self-disclosure score (or the self-disclosure score), assuming that the proposal for the current behavioral change support stage has been completed. That is, the behavioral change support stage determination unit 53 determines destination behavioral change support stage information based on the level of the self-disclosure score (or the self-disclosure score) from the behavioral support stage information. The behavior change support stage determination unit 53 then updates the behavior change support stage to the latest one by overwriting the information indicating the determined (transition destination) behavior change support stage (behavior change support stage information) in the behavior change support stage DB 74. The behavior change support stage determination unit 53 also outputs the determined (transition destination) behavior change support stage information to the proposal content selection unit 54.

[0088] S14: The proposal content selection unit 54 selects specific proposal content from the behavior tree (FIGS. 11A to 11H) according to the behavior change support stage after the determination (transition destination), either uniquely (FIGS. 11A, 11G), randomly taking priority into consideration (FIG. 11B), in numerical order (FIG. 11C), randomly (FIGS. 11D, 11F), or so as to conform to predetermined conditions using self-disclosure data (FIG. 11E). The proposal content selection unit 54 then outputs data of the selected specific proposal content to the proposal content transmission unit 55.

[0089] S15: The proposal content sending unit 55 sends data of the selected specific proposal content to the operation terminal 30. As a result, the output unit 32 of the operation terminal 30 displays one batch of the agent's specific proposal content (e.g., the question "What do you usually eat for breakfast?" and options indicating multiple answers to this question) as shown in FIG. 12. When the user Y selects one of the options, the data sending unit 31 sends the selected answer to the dialogue device 50, and the data acquiring unit 51 receives the answer, causing the process to proceed further from the above process S11. Again, in process S15, the output unit 32 displays one batch of the agent's proposal content (e.g., "What time do you usually eat breakfast?" and each option) as shown in FIG. 12.

[0090] In this way, the dialogue device 50 (agent) selects the content of a proposal to be made to the user Y in accordance with the depth of the relationship between the user Y and the dialogue device 50, which changes as the dialogue between the agent and the user Y progresses.

[0091] [Effects of this embodiment] As described above, according to this embodiment, the dialogue device 50 (agent) selects the content of the proposal to be made to the user Y in accordance with the depth of the relationship between the user Y and the dialogue device 50, which changes as the dialogue between the agent and the user Y progresses, thereby achieving the effect of making it easier for the user to accept the content of the proposal regarding behavioral change.

[0092] For example, when user Y first started to interact with the agent, the interaction device 50 (agent) suddenly started to say, 6-1As shown in Fig. 11F, even if the agent suggests, "It may be a little difficult, but if you incorporate some ideas, it will be easier to tackle," there is a high possibility that user Y will not trust the agent and will ignore it. However, as the dialogue between the agent and user Y progresses and the relationship between user Y and the dialogue device 50 deepens, user Y will 6-1 " is more likely to carry out behavioral modification. This can improve the significance of the suggestion to support behavioral modification for user Y.

[0093] [Supplementary Note] The present invention is not limited to the above-described embodiment, and may have the following configurations or processes (operations), for example.

[0094] (1) The interactive device 50 can be realized by a computer and a program, but the program can also be recorded on a (non-temporary) recording medium and provided, or the program can be provided via a communication network such as the Internet.

[0095] (2) The processor 1001, which is hardware, may be a single processor or multiple processors.

[0096] REFERENCE SIGNS LIST 10 Communication system 20 Mobile terminal (an example of a user terminal) 30 Operation terminal (an example of a user terminal) 50 Dialogue device (an example of a proposal device) 51 Data acquisition unit (an example of an acquisition unit) 52 Self-disclosure score calculation unit (an example of a calculation unit) 53 Behavioral change support stage determination unit (an example of a determination unit) 54 Proposal content selection unit (an example of a selection unit) 55 Proposal content transmission unit (an example of a transmission unit) 70 Database server 71 Self-disclosure data DB (an example of a self-disclosure data management unit) 72 User information DB (an example of a user information management unit) 73 Self-disclosure score DB (an example of a self-disclosure score management unit) 74 Behavioral change support stage DB (an example of a behavioral change support stage management unit) 75 Proposal scenario DB (an example of a proposal scenario management unit)

Claims

1. A proposal device that makes suggestions for behavior change to a user in an agent-based interactive format, comprising: a behavior change support stage management unit that manages behavior change support stage information indicating the behavior change support stage for the user based on the user's trust in the agent; a calculation unit that calculates a self-disclosure score, which is a value corresponding to the number of times a specified user has disclosed, based on self-disclosure data about the specified user; a determination unit that determines a destination behavior change support stage from the behavior change support stage information according to the self-disclosure score; a selection unit that selects a specific proposal content from a plurality of proposal contents corresponding to the behavior change support stage according to the destination behavior change support stage; and a transmission unit that transmits data of the specific proposal content to a specified user terminal of the specified user.

2. The proposal device according to claim 1, wherein the calculation unit calculates the self-disclosure score, which is a value corresponding to the disclosure rate of a specified user relative to the disclosure rate of all users based on the self-disclosure data.

3. The proposal device described in claim 1, wherein the determination unit determines the behavior change support stage to which the transition is to be made using a transition condition management table that associates and manages the current behavior change support stage, the transition conditions related to the self-disclosure score, and the behavior change support stage to which the transition is to be made.

4. The proposal device described in claim 1, wherein the selection unit randomly selects the specific proposal content from among a plurality of predetermined proposal contents when there are a plurality of predetermined proposal contents corresponding to the behavioral change support stage to which the transition is to be made.

5. The suggestion device described in claim 4, wherein when a plurality of predetermined suggestion contents corresponding to the behavioral change support stage to which the transition is to be made are divided into a plurality of categories, the selection unit selects a predetermined category according to the priority of the category and randomly selects a specific suggestion content within the predetermined category.

6. The suggestion device according to claim 1, wherein the transmission unit transmits to the user terminal, as the specific suggestion content, data of a question and options indicating a plurality of answers to the question.

7. A proposal method executed by a proposal device that makes suggestions for behavior change to a user in an agent-based interactive format, wherein the proposal device has a behavior change support stage management unit that manages behavior change support stage information that indicates the behavior change support stage for the user based on the user's trust in the agent, and the proposal device executes the following: a calculation process that calculates a self-disclosure score, which is a value corresponding to the number of times a specified user has disclosed, based on self-disclosure data about the specified user; a determination process that determines a behavior change support stage to which the specified user will transition based on the self-disclosure score from the behavior change support stage information; a selection process that selects a specific proposal content from a plurality of proposal contents corresponding to the behavior change support stage, based on the behavior change support stage to which the specified user will transition; and a transmission process that transmits data of the specific proposal content to a specified user terminal of the specified user.

8. A program for causing a computer to execute the method according to claim 7.

Images