Information management system, information management method, and program

Abstract

According to the present disclosure, when updating a content generation model that generates behavior modification content, consideration is given to whether or not it is necessary to acquire update information to be used for the update. This information processing system comprises: a contribution evaluation unit that evaluates a contribution, to an update of a content generation model that generates behavior modification content for causing a user to execute a predetermined target behavior, of update information for updating the content generation model, on the basis of a contribution evaluation rule for evaluating the contribution, to the update of the content generation model, of the update information; an acquisition necessity determination unit that determines, on the basis of the contribution, whether or not it is necessary to acquire the update information; an update information acquisition unit that acquires the update information, on the basis of the result of the determination of the necessity; and a model update unit that updates the content generation model using the acquired update information.

Description

Information Management System, Information Management Method, Program Cross - reference to Related Applications

[0001] This application is based on Japanese Patent Application No. 2024 - 163605 filed on September 20, 2024, the content of which is incorporated herein by reference.

[0002] This disclosure relates to an information management system, an information management method, and a program.

[0003] In recent years, efforts to apply an approach based on theories of behavioral science, which scientifically studies human behavior, to service development have spread in various fields such as public policy, medicine, retail, and education. Also, systems for technically realizing the support for behavior modification and habituation of target persons in such various fields have been studied (for example, Patent Document 1). In the system described in Patent Document 1, behavioral data including various data measured for the behaviors of a plurality of target persons is analyzed, and based on the results of the analysis of the behavioral data, a stage index, which is an index as a criterion for a plurality of stages gradually leading to a behavior as a habituation target, and each of the plurality of stages according to the stage index are defined, for each pair of adjacent stages, the gap between the two stages constituting the pair is identified, for each stage pair, at least one of the reason why the identified gap exists and the measure for causing a behavior modification for a target person belonging to the lower stage to transition to the higher stage is identified from relationship information in which the relationship between the gap and the reason / measure is defined, and it is described that processing regarding the reason / measure identified for each stage pair is executed.

[0004] Japanese Unexamined Patent Application Publication No. 2020 - 140596

[0005] Incidentally, content generation models that produce behavioral change content to encourage behavioral change may be updated as needed to produce more effective behavioral change content. In order to update content generation models to ones that can produce higher quality behavioral change content, update information used to update content generation models may be easily obtained. Easily obtaining update information may compromise user privacy and may also incur significant costs for obtaining and managing update information.

[0006] Therefore, this disclosure aims to consider whether it is necessary to acquire update information used for updating content generation models that generate behavioral change content.

[0007] An information management system according to one aspect of this disclosure includes: a contribution evaluation unit that evaluates the contribution of update information to updating a content generation model based on a contribution evaluation rule that evaluates the contribution of update information to updating a content generation model, which generates behavior change content that causes a user to perform a predetermined target action; an acquisition necessity determination unit that determines whether or not to acquire the update information based on the contribution; an update information acquisition unit that acquires the update information based on the result of the necessity determination; and a model update unit that updates the content generation model using the acquired update information.

[0008] According to this disclosure, when updating a content generation model that generates behavioral change content, it is possible to consider whether or not it is necessary to acquire update information used for the update.

[0009] This figure shows an overview of the processing in the information processing system 100, which is one embodiment of the present invention. This figure shows an example of BCT classification. This figure shows the configuration of the information processing system 100, which is one embodiment of the present invention. This flowchart shows an example of the processing of the information processing system 100 in the first embodiment. This flowchart shows an example of the processing of the information processing system 100 in the second embodiment. This figure shows an example of the hardware configuration of the computer 600.

[0010] Embodiments of the present invention will be described with reference to the attached drawings. Figure 1 is a diagram showing an overview of the processing in an information processing system 100, which is one embodiment of the present invention.

[0011] As a first embodiment, the information processing system 100 determines whether or not to acquire the first user information based on the contribution according to a contribution evaluation rule that evaluates the degree of contribution of the first user information to the generation of behavioral change content.

[0012] In this case, first, the information processing system 100 receives an application from the information processing device 200 for the provision of behavioral change content and also receives the provision of first user information (S111). The information processing system 100 inputs the first user information into the content generation model and generates behavioral change content (S112). The information processing system 100 evaluates the contribution of the first user information to the generation of behavioral change content based on the contribution evaluation rule (S113). Based on the contribution, the information processing system 100 decides whether or not to acquire the first user information (S114).

[0013] Then, based on the result of the determination of necessity, the information processing system 100 obtains from the information processing device 200 the first user information that has been determined to be necessary to obtain when providing the behavioral change content (S115). The information processing system 100 inputs the first user information into the content generation model to generate the behavioral change content and provides it to the information processing device 200 (S116).

[0014] In this way, the information processing system 100 can reduce the amount of user information it acquires from users and avoid the indiscriminate acquisition of user information. The indiscriminate acquisition of user information includes, for example, acquiring user information that has little effect on generating high-quality behavioral change content. By avoiding the indiscriminate acquisition of user information, user privacy can be protected, and the entity that acquires user information (for example, the administrator of the information processing system 100) can reduce the costs of acquiring user information and the costs of properly managing user information.

[0015] Furthermore, as a second embodiment, the information processing system 100 determines whether or not to acquire update information for updating the content generation model based on the contribution of the update information according to a contribution evaluation rule that evaluates the contribution of the update information to the content generation model, and updates the content generation model based on the acquired update information.

[0016] In this case, first, the information processing system 100 receives update information from the information processing device 200 (S121). Based on the contribution evaluation rules, the information processing system 100 evaluates the contribution of the update information to the update of the content generation model (S122). Based on the contribution, the information processing system 100 decides whether or not to acquire the update information (S123).

[0017] Then, when updating the content generation model, the information processing system 100 obtains update information from the information processing device 200 (S124). Based on the update information, the information processing system 100 updates the content generation model (S125).

[0018] In this way, the information processing system 100 can reduce the amount of update information it acquires when updating the content generation model, and can avoid the easy acquisition of update information. By avoiding the easy acquisition of update information, the privacy of users corresponding to the update information can be protected, and the entity that acquires the update information (for example, the administrator of the information processing system 100) can reduce the costs of acquiring the update information and the costs of properly managing the update information.

[0019] Note that the two information processing devices 200 shown in Figure 1 may be different information processing devices or the same information processing device.

[0020] In this embodiment, behavior change content is content that causes the user to perform a predetermined target behavior (also called behavior change). That is, behavior change content is, for example, content that has the effect of changing the user's behavior. In this embodiment, the case in which the behavior change content is a single behavior change content will be described as an example, but if the behavior change content is a series of behavior change content, the behavior change content can be appropriately reinterpreted as a series of behavior change content.

[0021] Behavior change content is, for example, content generated based on behavior change techniques (BCTs). For example, according to BCTTv1 (Michie S, Richardson M, Johnston M, et al.: The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Ann Behav Med 2013; 46: 81-95.), 93 BCTs in 16 groups are defined. However, the definition of BCTs is not limited to BCTTv1; they may be defined in any way as long as they comprehensively cover behavior change techniques.

[0022] BCTTv1 defines 16 BCT groups: 1. Goals and planning, 2. Feedback and monitoring, 3. Social support, 4. Shaping knowledge, 5. Natural consequences, 6. Comparison of behavior, 7. Associations, 8. Repetition and substitution, 9. Comparison of outcomes, 10. Reward and threat, 11. Regulation, 12. Antecedents, 13. Identity, 14. Scheduled consequences, 15. Self-belief, and 16. Covert learning.

[0023] In BCTTv1, each of the 16 BCT groups shown in Figure 2 contains one or more BCTs. For example, BCT group "5. Natural consequences" contains BCTs such as "5.5. Anticipated regret." Similarly, BCT group "10. Reward and threat" contains BCTs such as "10.11. Future punishment." Although not shown in the figures, other groups also contain one or more BCTs.

[0024] Furthermore, each BCT has components, and the degree to which each BCT is contained in a given content may be shown as a component value. In addition, the sum of the component values ​​of each BCT belonging to the same BCT group may be shown as the component value of the BCT group.

[0025] Behavioral changes promoted by behavioral change content may include, but are not limited to, language learning, dieting, purchasing financial products offered by financial institutions, regular medical checkups, or use of public services. Furthermore, behavioral change content is not limited to being provided to users by electronic means such as applications installed on their devices, web pages, emails, or short messages, but may also be provided to users by other means (e.g., customer service, postal mail). Moreover, behavioral change content may include, but is not limited to, text information related to a specific service (e.g., chat logs), videos, still images, audio of conversations, or application data, and may include any information related to the content of the behavioral change.

[0026] In generating and evaluating behavioral change content, a coordinate system is created with multiple behavioral change factors as axes, and a target coordinate or target area (hereinafter collectively referred to as the target area) is set within this coordinate system. Whether or not a user has undergone behavioral change is then evaluated by whether or not the user's coordinate in the coordinate system has approached the target area. Hereafter, the distance calculated in this coordinate system will be referred to as the "Behavioral Scientific Distance (BSD)".

[0027] In this coordinate system, by setting the current coordinate and the target area that the intervention should aim for for each user, the behavioral science distance from the current coordinate to the target area can be represented by a multidimensional vector, and the Behavioral Change Therapy (BCT) that should be adopted to bridge this distance can be visualized objectively and clearly. In other words, higher quality behavioral change content is, for example, behavioral change content that further reduces the behavioral science distance from the current coordinate to the target area.

[0028] Behavioral change factors are, for example, factors that motivate users to take target actions. Multiple behavioral change factors in a target field can be identified based on academic surveys or opinion surveys, theories of behavioral science, user persona development, or behavioral process maps. Examples of behavioral change factors include "Capacity," "Opportunity," and "Motivation" in the COM-B model, which is commonly used in behavioral science. Alternatively, as another example of behavioral change factors, the factors defined in the Integrated Behavioral Model (IBM) (also called "IBM factors") include experiential attitude, instrumental attitude, injunctive norm, descriptive norm, perceived control, self-efficacy, knowledge, skills, salience of the behavior, environmental constraints, and habit. In this embodiment, behavioral change factors including at least two of these may be identified.

[0029] The IBM factors described above are merely examples, and higher-level factors encompassing the IBM factors may be identified as behavioral change factors. For example, the empirical attitude and instrumental attitude may be encompassed under "Attitude," the referential and descriptive norms under "Perceived norm," the sense of control over behavior and self-efficacy under "Personal Agency," the knowledge and skills under "Knowledge," the importance of behavior under "Importance," and environmental constraints under "Friction." Alternatively, lower-level factors obtained by dividing the IBM factors may be identified as behavioral change factors. In this embodiment, any factors relating to human behavioral will may be adopted as behavioral change factors, such as factors defined in the COM-B model or behavioral models other than IBM.

[0030] The ease with which users respond to behavioral change factors can vary from person to person. That is, some users may be more likely to change their behavior through behavioral change content that strongly influences a first behavioral change factor, while other users may be more likely to change their behavior through behavioral change content that strongly influences a second behavioral change factor. Furthermore, even within the same user, the ease with which they respond to behavioral change factors can vary depending on the passage of time, the time and environment in which the behavioral change content is presented, and the target behavior and the content of the behavioral change. This ease with behavioral change factors is sometimes described as a user's behavioral characteristic. Thus, the behavioral scientific distance at which behavioral change content can exert its effect can vary depending on the user's behavioral characteristics.

[0031] Figure 3 shows the configuration of an information processing system 100, which is one embodiment of the present invention. The information processing system 100 is connected to the information processing device 200 via a network such as the Internet. Details of the information processing system 100 will be described later.

[0032] The information processing device 200 outputs at least one of user information relating to the user and update information for updating the content generation model to the information processing system 100. The information processing device 200 may be, for example, a computer, smartphone, tablet terminal, personal computer, etc.

[0033] The information processing device 200 may be a user device used by the user. In this case, the information processing device 200 does not need to output user information to the information processing system 100, but it does not need to output update information.

[0034] Furthermore, the information processing device 200 may also be an information management device for managing information. In this case, the information processing device 200 may output update information to the information processing system 100, but may not output user information.

[0035] Although Figure 3 shows one information processing device 200, the information processing device 200 may be multiple information processing devices 200 (for example, a user device and an information management device).

[0036] Next, the details of the information processing system 100 will be described. The information processing system 100 includes a storage unit 110, an acquisition processing unit 120, an update information acquisition unit 130, a behavioral characteristics acquisition unit 140, a content generation unit 150, a content evaluation unit 160, a contribution evaluation unit 170, an acquisition necessity determination unit 180, a content output unit 190, and a model update unit 195. Each of the units shown in Figure 3 can be realized, for example, by using a storage area or by having a processor execute a program stored in the storage area.

[0037] The memory unit 110 stores information processed by the information processing system 100. The memory unit 110 can store, for example, user information, update information, behavioral characteristic information, behavioral change content, effect evaluation result information, contribution evaluation result information, and decision result information, which will be described later.

[0038] The acquisition processing unit 120 acquires user information and stores the acquired user information in the storage unit 110. The acquisition processing unit 120 acquires information from, for example, the information processing device 200.

[0039] Before the acquisition necessity determination process by the acquisition necessity determination unit 180 described later, the acquisition processing unit 120 can acquire user information (for example, first user information) that is the target of the acquisition necessity determination process.

[0040] Based on the result of the necessity determination by the acquisition necessity determination unit 180, the acquisition processing unit 120 can perform an acquisition process of acquiring user information (for example, first user information). Further, based on the result of the necessity determination by the acquisition necessity determination unit 180, the acquisition processing unit 120 can perform an acquisition process of not acquiring user information (for example, first user information).

[0041] User information includes, but is not limited to, biometric information related to the user's body, position information of the user or a device (information processing device 200) owned by the user, or device information related to the device, and information related to the content of the answer results of the questionnaire. Biometric information includes, for example, information related to the content of the results of the user's health examination, and vital information such as the user's heartbeat, but is not limited thereto.

[0042] Before the acquisition necessity determination process, the user information acquired as the target of the acquisition necessity determination process may be user information having a larger amount of information (at least either the content or the items) than the user information acquired based on the result of the necessity determination. That is, the information processing system 100 can, so to speak, first widely acquire available user information and then further acquire only the user information determined to be necessary for acquisition, and generate action-variable content provided to the user.

[0043] The update information acquisition unit 130 acquires update information and stores the acquired update information in the storage unit 110.

[0044] Before the acquisition necessity determination process by the acquisition necessity determination unit 180 described later, the update information acquisition unit 130 can acquire update information that is the target of the acquisition necessity determination process.

[0045] The update information acquisition unit 130 can perform an acquisition process of acquiring update information based on the result of the necessity determination by the acquisition necessity determination unit 180. Further, the update information acquisition unit 130 can perform an acquisition process of not acquiring update information based on the result of the necessity determination by the acquisition necessity determination unit 180.

[0046] The update information may be any information that can update the content generation model. The update information may include user information, and may also include the behavior characteristic information described later.

[0047] Before the acquisition necessity determination process, the update information acquired as the target of the acquisition necessity determination process may be update information with a larger amount of information (at least any one of content, type, and items) than the update information acquired based on the result of the necessity determination. That is, the information processing system 100 can, so to speak, first widely acquire the acquirable update information, and then further acquire only the update information determined to be necessary for acquisition to update the content generation model.

[0048] The behavior characteristic acquisition unit 140 acquires behavior characteristic information regarding the behavior characteristics of the user, and stores the acquired behavior characteristic information in the storage unit 110. Here, the behavior characteristic is, for example, the ease of reaction of each behavior change factor in the user.

[0049] The behavior characteristic information of the same user may be different each time the behavior characteristic acquisition unit 140 acquires it. That is, the behavior characteristic information may be different for the same user according to the passage of time, the situation such as the time when the behavior characteristic acquisition unit 140 acquires and the environment at that time, the target behavior, and the content of the behavior change.

[0050] The content generation unit 150 generates behavior change content based on a content generation model that generates behavior change content for causing the user to execute a predetermined target behavior, and stores the generated behavior change content in the storage unit 110.

[0051] The content generation model may be, for example, a model that generates behavioral change content in response to user information input. In this case, the content generation unit 150 can input first user information into the content generation model to generate behavioral change content (behavioral change content after input). The content generation unit 150 may also generate behavioral change content (behavioral change content before input) without inputting first user information into the content generation model. In this embodiment, generating behavioral change content (behavioral change content before input) without inputting first user information means inputting user information that does not include the first user information, that is, user information different from the first user information, into the content generation model to generate behavioral change content (behavioral change content before input).

[0052] Furthermore, the content generation model may be a model that generates behavioral change content based on the user's behavioral characteristics in response to the input of user information and behavioral characteristic information. In other words, in this case, the content generation unit 150 can input user information and behavioral characteristic information into the content generation model and generate behavioral change content based on the user's behavioral characteristics.

[0053] Furthermore, the content generation model may be a model that generates a series of behavioral change content corresponding to each of the multiple actions performed up to a predetermined target action. In other words, in this case, the content generation unit 150 can generate a series of behavioral change content based on the content generation model.

[0054] Furthermore, the content generation unit 150 can generate behavioral change content (post-updated behavioral change content) based on the updated content generation model, which is updated by the model update unit 195, using the pre-update content generation model before it is updated by the model update unit 195, which will be described later.

[0055] The content generation model is, for example, a model that generates behavioral change content that has the effect of shortening the behavioral science distance from the current coordinates to the target area.

[0056] The content generation model may generate content using, for example, artificial intelligence (AI) technology, particularly generative AI. Specifically, when the content generation model receives user information as input, it generates generation request information (so to speak, a prompt) that incorporates the user information into pre-set generation request information, obtains behavioral change content generated by the generative AI using the generated generation request information, and outputs the behavioral change content. Here, the generation request information may include, for example, information about the target behavior or the content of the behavioral change, or information about the type of behavioral change content to be generated. Furthermore, the generation request information may be generation request information that causes the generative AI to generate behavioral change content that has the greatest effect in reducing the behavioral scientific distance from the current coordinates to the target area.

[0057] Furthermore, the content generation model may be a model that generates behavioral change content by, for example, selecting at least one behavioral change content from among several pre-configured behavioral change content based on user information. In this case, the content generation model may select the behavioral change content that has the greatest effect in reducing the behavioral scientific distance from the current coordinate to the target area.

[0058] Furthermore, the content generation model may, for example, generate a single behavioral change content by combining selected behavioral change content or elements of behavioral change content from among several pre-configured behavioral change content or elements of behavioral change content.

[0059] Furthermore, the content generation model may generate a series of behavioral change content pieces that consist of multiple behavioral change content pieces.

[0060] The content evaluation unit 160 evaluates the effectiveness of behavior change content based on content evaluation rules that evaluate the effectiveness of behavior change content in getting users to perform predetermined target behaviors, and stores the effectiveness evaluation result information regarding the evaluated effect in the storage unit 110.

[0061] Content evaluation rules may, for example, be correspondence information that shows the relationship between multiple elements for evaluating content that are set in advance and the weight of each element. Alternatively, content evaluation rules may be evaluation models that take content as input and output a score that indicates the evaluation result of that content.

[0062] Furthermore, the content evaluation rule may also evaluate the magnitude of the behavioral scientific distance over which behavioral change content acts. In other words, in this case, the content evaluation rule may be one that gives a higher evaluation to behavioral change content that more effectively reduces the behavioral scientific distance to the user's target behavior. The content evaluation rule may also be one in which the behavioral scientific distance over which behavioral change content acts is correlated with the evaluation result.

[0063] Furthermore, content evaluation rules may, for example, evaluate content (e.g., text) as low quality if its content is general, and evaluate content (e.g., text) as high quality if its content is distinctive. Here, whether or not the content is distinctive may be evaluated based on, for example, the proportion of proper nouns contained in the content; that is, the higher the proportion of proper nouns contained in the content, the more distinctive the content may be evaluated.

[0064] Furthermore, the content evaluation rules may also stipulate that, for example, content composed of multiple sentences may be evaluated as high quality if the similarity in meaning between each sentence is low and the information density is high, if the ratio of kanji and hiragana usage is appropriate and the text is highly readable, if the content is contextually appropriate even when translated into other languages, or if the content includes fonts, sizes, graphics, or colors that take universal design into consideration.

[0065] The content evaluation unit 160 can output evaluation results based on the content evaluation rules, such as a quantitative score expressed using numerical values ​​(e.g., a score corresponding to behavioral science distance) or a qualitative score (e.g., "Excellent," "Good," etc.).

[0066] The content evaluation unit 160 can evaluate the effectiveness of post-input behavior change content based on content evaluation rules. The content evaluation unit 160 can also evaluate the effectiveness of pre-input behavior change content based on content evaluation rules. Furthermore, the content evaluation unit 160 can evaluate the effectiveness of a series of post-input behavior change content generated by inputting first user information into the content generation model.

[0067] The content evaluation unit 160 can evaluate the effectiveness of the updated behavior change content based on the content evaluation rules. Furthermore, the content evaluation unit 160 can evaluate the effectiveness of the behavior change content (pre-update behavior change content) generated based on the pre-update content generation model, also based on the content evaluation rules. Additionally, the content evaluation unit 160 can evaluate the effectiveness of a series of updated behavior change content.

[0068] In this way, the content evaluation unit 160 can evaluate how important the generated behavioral change content is for achieving the target behavior.

[0069] The contribution evaluation unit 170 evaluates the contribution of the first user information to the generation of behavioral change content based on a first contribution evaluation rule that evaluates the contribution of the first user information to the generation of behavioral change content, and stores the contribution evaluation result information regarding the evaluated contribution in the storage unit 110.

[0070] The contribution evaluation unit 170 can evaluate the contribution of the first user information to the effect evaluated by the content evaluation unit 160.

[0071] Here, the content evaluation unit 160 may evaluate the magnitude of the behavioral science distance over which the behavioral change content acts as the effect of the behavioral change content.

[0072] The contribution evaluation unit 170 can evaluate the contribution of the first user information based on the effects of the pre-input behavior change content and the post-input behavior change content, which have been evaluated by the content evaluation unit 160. In other words, in this case, the first contribution evaluation rule may be, for example, a rule that evaluates the contribution of the first user information based on the effects of the pre-input behavior change content and the effects of the post-input behavior change content.

[0073] Specifically, first, the content evaluation unit 160 evaluates the effect of the pre-input behavior change content and outputs pre-input effect evaluation result information as an evaluation result. Next, the content evaluation unit 160 evaluates the effect of the post-input behavior change content and outputs post-input effect evaluation result information as an evaluation result. Then, the contribution evaluation unit 170 compares the pre-input effect evaluation result information and the post-input effect evaluation result information to evaluate the contribution of the first user information. At this time, the contribution evaluation unit 170 may, for example, evaluate the difference or discrepancy between the effect (e.g., score) shown by the pre-input effect evaluation result information and the effect (e.g., score) shown by the post-input effect evaluation result information as the contribution of the first user information. In this way, the contribution evaluation unit 170 can evaluate the contribution of further investment of the resource of the first user information to the return of the evaluation result, so to speak, from the perspective of ROI (Return On Investment). In other words, the contribution evaluation unit 170 can evaluate the contribution according to the degree to which it is expected that more effective behavior change content can be generated when the first user information is acquired. In this case, the contribution evaluation unit 170 can evaluate the degree of contribution from the perspective of how much the first user information contributes to determining behavioral change content that is likely to enable the user to achieve the target behavior.

[0074] In this case, the difference between the effect (e.g., score) shown by the pre-input effect evaluation result information and the effect (e.g., score) shown by the post-input effect evaluation result information may be the difference in behavioral science distance. Let's consider an example where the target behavior is going to work, and the first user information is that the user is at home. In this case, for example, the post-input behavior change content may be a message such as, "Why don't you leave the house?". The pre-input behavior change content may be a message such as, "Why don't you go to the office?" rather than, "Why don't you leave the house?", because the user's location is unknown. In this case, the behavioral science distance at which the message "Why don't you leave the house?" acts may be greater than the behavioral science distance at which the message "Why don't you go to the office?" acts. This is because, for example, the word "office" may be a barrier to the user's behavior change. In such a case, the contribution evaluation unit 170 can evaluate, for example, the difference in behavioral science distance as the contribution of the first user information that the user is at home. In this way, the contribution evaluation unit 170 can evaluate to what extent the first user information contributes when generating effective behavior change content from multiple anticipated behavior change contents.

[0075] Furthermore, the contribution evaluation unit 170 can evaluate the contribution of the first user information based on external environment information relating to the external environment associated with the execution of a predetermined target action. In other words, in this case, the first contribution evaluation rule may be, for example, a rule that evaluates the contribution of the first user information based on external environment information.

[0076] In this case, the contribution evaluation unit 170 can evaluate, so to speak, the extent to which the first user information contributes to responding to changes in the environment surrounding the target behavior. Let's consider an example where the target behavior is commuting to work, the first user information is the route the user uses to commute, and the external environment information is the operating status of the train used for commuting. In this case, when the train used for commuting is running normally, the influence of the first user information on the generation of behavior change content and the achievement of the target behavior is small, but when the train used for commuting is not running normally, the influence of the first user information on the generation of behavior change content and the achievement of the target behavior may become large. This is because, for example, when the train used for commuting is not running normally, the user may be reluctant to go to work. When the train used for commuting is not running normally, instead of or in addition to the message "Why don't you leave home?", behavior change content that suggests the time until the train schedule used by the user to commute returns to normal or an alternative route different from the route the user uses to commute may be effective. In other words, the first user information, such as the route a user uses for commuting, which has a small contribution during normal train schedules, may become the first user information with a high contribution during train schedule abnormalities. In this way, the contribution evaluation unit 170 can evaluate the contribution by further considering external environmental information.

[0077] External environmental information may include, but is not limited to, information about changes in the environment surrounding the target behavior, such as news reports about misconduct at the user's workplace.

[0078] The contribution evaluation unit 170 evaluates the contribution of the update information to the content generation model based on a second contribution evaluation rule that evaluates the contribution of the update information to the content generation model, and stores the contribution evaluation result information regarding the evaluated contribution in the storage unit 110.

[0079] The contribution evaluation unit 170 can evaluate the contribution of update information to the content generation model based on the effect of the pre-update behavior change content and the effect of the post-update behavior change content. In other words, in this case, the second contribution evaluation rule may be a rule that evaluates the contribution based on the effect of the pre-update behavior change content and the effect of the post-update behavior change content. Here, the information input to the content generation model may be the same for the pre-update behavior change content and the post-update behavior change content. In other words, the difference between the pre-update behavior change content and the post-update behavior change content may be a difference in the content generation model, or in other words, a difference between the pre-update content generation model and the post-update content generation model.

[0080] Specifically, first, the content evaluation unit 160 evaluates the effect of the pre-update behavior change content and outputs pre-update effect evaluation result information as an evaluation result. Next, the content evaluation unit 160 evaluates the effect of the post-update behavior change content and outputs post-update effect evaluation result information as an evaluation result. Then, the contribution evaluation unit 170 compares the pre-update effect evaluation result information and the post-update effect evaluation result information and evaluates the contribution of the update information to the content generation model. At this time, the contribution evaluation unit 170 may, for example, evaluate the difference or discrepancy between the effect (e.g., score) shown by the pre-update effect evaluation result information and the effect (e.g., score) shown by the post-update effect evaluation result information as the contribution of the update information to the content generation model. In this way, the contribution evaluation unit 170 can evaluate the contribution of further input of the resource of update information to the return of updating (so to speak, improving) the content generation model, so to speak, from the perspective of ROI (Return On Investment). In other words, the contribution evaluation unit 170 can evaluate the degree of contribution in relation to the expectation that the pre-update content generation model can be updated to a post-update content generation model capable of generating more effective behavioral change content when update information is acquired. In this case, the contribution evaluation unit 170 can evaluate the degree of contribution from the perspective of how much the update information contributes to updating the content generation model that generates behavioral change content that is highly likely to help users achieve their target behavior.

[0081] In this case, the difference between the effect (e.g., score) shown by the pre-update effect evaluation result information and the effect (e.g., score) shown by the post-update effect evaluation result information may be the difference in behavioral science distance. In this way, the contribution evaluation unit 170 can evaluate to what extent the update information contributes when updating the pre-update content generation model to the post-update content generation model that can generate more effective behavior change content. For example, if the target behavior is going to work, it is conceivable to register the user's workplace (e.g., the user's company) as update information. By registering the user's company, the behavior change content can include specific station names related to the workplace (e.g., the nearest station to the workplace). If the pre-update behavior change content does not include station names, but the post-update behavior change content does, the system may switch whether to acquire workplace information as update information from the user depending on whether the post-update evaluation result is higher than the pre-update evaluation result.

[0082] Furthermore, the contribution evaluation unit 170 can evaluate the contribution of update information to the content generation model based on external environmental information. In other words, in this case, the second contribution evaluation rule may be, for example, a rule that evaluates the contribution of update information to the content generation model based on external environmental information. For example, suppose the target behavior is going to work, and the user has registered information about the company they belong to. In this case, the location where the user goes to work may change due to office relocation, etc. By obtaining information about the relocation of the user's company as external environmental information, it may be better to change the station names, etc., included in the behavior change content sent to the user.

[0083] The acquisition necessity determination unit 180 determines whether or not to acquire the first user information based on the contribution evaluated according to the first contribution evaluation rule, and stores the decision result information regarding the decision result in the storage unit 110.

[0084] Furthermore, the acquisition necessity determination unit 180 determines whether or not to acquire the update information based on the contribution evaluated according to the second contribution evaluation rule, and stores the decision result information regarding the decision result in the storage unit 110.

[0085] The decision result information may, for example, be information associated with each of the first user information and update information, indicating whether acquisition is required ("Required") or not required ("Not Required").

[0086] The acquisition necessity determination unit 180 may, for example, determine that acquisition is necessary if the contribution exceeds a predetermined threshold.

[0087] Thus, the acquisition necessity determination unit 180 can determine, for example, that it is necessary to acquire first user information that is expected to enable the generation of more effective behavioral change content. Furthermore, the acquisition necessity determination unit 180 can determine that it is necessary to acquire update information that is expected to enable the updating of a more effective content generation model.

[0088] The content output unit 190 outputs behavioral change content obtained by inputting the first user information, which was acquired based on the acquisition necessity determination unit 180, into the content generation model. As a result, the information processing system 100 can output behavioral change content that is generated based on the first user information, which is expected to generate more effective behavioral change content.

[0089] The content output unit 190 may output behavior change content to, for example, an information processing device 200. Alternatively, the content output unit 190 may output behavior change content to an information processing device used by a person who provides behavior change content to a user.

[0090] The model update unit 195 updates the content generation model using the update information. The model update unit 195 can update the pre-update content generation model, which is before it is updated by the update information, using the update information, and generate the post-update content generation model.

[0091] The model update unit 195 can update the content generation model using the update information acquired based on the acquisition necessity determination unit 180.

[0092] Furthermore, the model update unit 195 can update the content generation model based on the contribution evaluated by the contribution evaluation unit 170 based on the second contribution evaluation rule. In this case, the model update unit 195 may update the content generation model using update information having a contribution exceeding a predetermined threshold. This allows the information processing system 100 to update the content generation model without needing to acquire further update information, thereby simplifying information processing. Moreover, in this case, when the content generation model is updated next time, the update information acquisition unit 130 may acquire only the update information that has been determined to be necessary, and the processing of the second embodiment in the information processing system 100 may be performed. This allows the update information to be acquired to be selected sequentially.

[0093] Furthermore, the model update unit 195 can update the content generation model using behavioral characteristic information.

[0094] The first user information may consist of multiple pieces of first user information. In this case, the contribution evaluation unit 170 can evaluate the contribution of each of the multiple pieces of first user information based on the first contribution evaluation rule. The acquisition necessity determination unit 180 determines whether or not to acquire each of the multiple pieces of first user information based on the contribution of each piece of first user information, and the acquisition processing unit 120 can acquire the first user information that has been determined to be necessary from among the multiple pieces of first user information.

[0095] The update information may consist of multiple pieces of update information. In this case, the contribution evaluation unit 170 can evaluate the contribution of each of the multiple pieces of update information. The acquisition necessity determination unit 180 determines whether each of the multiple pieces of update information needs to be acquired, and the update information acquisition unit 130 can acquire the update information that has been determined to need to be acquired from among the multiple pieces of update information.

[0096] Next, a specific example of the information processing system 100 will be described.

[0097] Figure 4 is a flowchart showing an example of the processing of the information processing system 100 in the first embodiment.

[0098] First, the acquisition processing unit 120 acquires multiple user information (S401). Here, the multiple user information may be, for example, user information of a common type for multiple users (for example, multiple users belonging to a predetermined group). Next, the content generation unit 150 generates behavior change content based on the multiple user information, and the content evaluation unit 160 evaluates the generated behavior change content (S402). At this time, the content generation unit 150 may generate behavior change content in which all of the multiple user information is input into the content generation model, and behavior change content in which a part of the multiple user information (for example, the user information included in each of the multiple user information is excluded) is input into the content generation model. The content generation unit 150 may also generate behavior change content by further inputting behavior characteristic information acquired by the behavior characteristic acquisition unit 140 into the content generation model. The contribution evaluation unit 170 evaluates the contribution of each of the multiple user information based on the first contribution evaluation rule (S403). At this time, the contribution evaluation unit 170 may, for example, evaluate the contribution of each of the multiple user information pieces based on the effect of the post-input behavior change content and the evaluation of the pre-input behavior change content. The acquisition necessity determination unit 180 determines whether or not to acquire each of the multiple user information pieces (S404).

[0099] Then, the acquisition processing unit 120 performs an acquisition process to acquire user information (for example, first user information) from among the multiple user information based on the result of the necessity determination, and performs an acquisition process to decide not to acquire user information (for example, second user information) from among the multiple user information (S405). At this time, the types of user information to be acquired and the types of user information not to be acquired may differ for each user. The content generation unit 150 generates behavior change content based on the user information acquired based on the necessity determination (for example, first user information), and the content output unit 190 outputs the behavior change content (S406).

[0100] Thus, the information processing system 100 first acquires multiple user information, determines whether or not to acquire each of the multiple user information based on the contribution of each of the multiple user information, and acquires the user information that is deemed necessary to acquire (for example, the first user information). That is, for example, when the information processing system 100 provides behavioral change content based on user information, it acquires multiple user information before providing the behavioral change content, identifies the user information that needs to be acquired in order to provide effective behavioral change content, and acquires the user information that is deemed necessary to acquire (for example, the first user information) when generating and providing the behavioral change content. This reduces the amount of user information acquired from users. As a result of reducing the amount of user information acquired from users, user privacy can be protected, and the entity that acquired the user information (for example, the administrator of the information processing system 100) can reduce the costs of acquiring user information and the costs of properly managing user information. Furthermore, the information processing system 100 may discard information, including information acquired in the past, regarding user information that is deemed unnecessary to acquire. As a result, the information processing system 100 can avoid the need to manage unnecessary information, reduce information management costs, and lower the risk of information leakage.

[0101] Figure 5 is a flowchart showing an example of the processing of the information processing system 100 in the second embodiment.

[0102] First, the update information acquisition unit 130 acquires multiple update information (S501). The contribution evaluation unit 170 evaluates the contribution of each of the multiple update information to the update of the content generation model based on the second contribution evaluation rule (S502). The acquisition necessity determination unit 180 determines whether or not to acquire each of the multiple update information (S503).

[0103] Then, the update information acquisition unit 130 acquires the update information that it has determined to be necessary to acquire from among the multiple update information based on the result of the necessity determination (S504). The model update unit 195 updates the content generation model using the update information that it has determined to be necessary to acquire (S505). At this time, the information processing system 100 may discard information that has been previously acquired, including the update information that it has determined to be unnecessary to acquire.

[0104] Next, with reference to Figure 6, an example of a hardware configuration when the information processing system 100 is implemented using a computer 600 will be described. Figure 6 is a diagram showing an example of the hardware configuration of the computer 600.

[0105] As shown in Figure 6, the computer 600 includes, for example, a processor 601, memory 602, storage device 603, input I / F unit 604, data I / F unit 605, communication I / F unit 606, and display device 607.

[0106] Computer 600 may be, for example, a server computer, a personal computer (e.g., desktop, laptop, tablet, etc.), a media computer platform (e.g., cable, satellite set-top box, digital video recorder, etc.), a handheld computer device (e.g., PDA, email client, etc.), or another type of computer or communication platform.

[0107] The processor 601 is a control unit that controls various processes in the computer 600 by executing programs stored in the memory 602.

[0108] Memory 602 is a storage medium such as RAM (Random Access Memory). Memory 602 temporarily stores the program code of the program executed by the processor 601, as well as data required during program execution.

[0109] The storage device 603 is a non-volatile storage medium such as a hard disk drive (HDD) or flash memory. The storage device 603 stores the operating system and various programs for realizing the above configurations.

[0110] The input interface unit 604 is a device for receiving input from the user. The input interface unit 604 may be, for example, a keyboard, mouse, touch panel, various sensors, or wearable devices. The input interface unit 604 may be connected to the computer 600 via an interface such as USB (Universal Serial Bus).

[0111] The data I / F unit 605 is a device for inputting data from outside the computer 600. The data I / F unit 605 is, for example, a drive device for reading data stored on various storage media. The data I / F unit 605 may be provided outside the computer 600. If the data I / F unit 605 is provided outside the computer 600, it is connected to the computer 600 via an interface such as USB.

[0112] The communication interface unit 606 is a device for performing data communication with external devices of the computer 600 via a network such as the Internet, either by wire or wireless connection. The communication interface unit 606 may be located outside the computer 600. If the communication interface unit 606 is located outside the computer 600, it is connected to the computer 600 via an interface such as USB.

[0113] The display device 607 is a device for displaying various types of information. The display device 607 may be, for example, a liquid crystal display, an organic EL (Electro-Luminescence) display, or a display for a wearable device. The display device 607 may be provided outside the computer 600. If the display device 607 is provided outside the computer 600, it is connected to the computer 600, for example, via a display cable. Also, if a touch panel is used as the input I / F unit 604, the display device 607 may be configured to be integrated with the input I / F unit 604.

[0114] One embodiment of the present invention has been described above. Based on the second contribution evaluation rule, the information processing system 100 evaluates the contribution of update information to the content generation model, determines whether or not to acquire update information based on the contribution, acquires update information based on the result of the determination of necessity, and can update the content generation model using the acquired update information. As a result, when updating a content generation model that generates behavioral change content, the information processing system 100 can consider whether or not to acquire update information used for the update.

[0115] Furthermore, the information processing system 100 can acquire behavioral characteristics information and update the content generation model using this information. This allows the information processing system 100 to update the content generation model while taking into account the user's behavioral characteristics.

[0116] Furthermore, the information processing system 100 can evaluate the effect of the updated behavioral change content obtained from the updated content generation model, which has been updated using update information, based on the content evaluation rules, and can evaluate the contribution of the update information to the update of the content generation model to the evaluated effect. As a result, the information processing system 100 can consider whether or not it is necessary to acquire update information based on the effect of the updated behavioral change content.

[0117] Furthermore, the information processing system 100 can further evaluate the effectiveness of the pre-update behavior change content based on content evaluation rules, and evaluate the contribution of update information to the update of the content generation model based on the effectiveness of the pre-update behavior change content and the effectiveness of the post-update behavior change content. As a result, the information processing system 100 can consider, for example, whether or not to acquire user information based on the difference or discrepancy between the effectiveness of the post-update behavior change content and the pre-update behavior change content, in a sense from the perspective of ROI.

[0118] Furthermore, the information processing system 100 can evaluate the contribution of update information to updating the content generation model based on external environmental information. This allows the information processing system 100 to consider whether or not it is necessary to acquire update information based on external environmental information.

[0119] Furthermore, the information processing system 100 can generate a series of behavioral change content and evaluate the effectiveness of the updated behavioral change content. This allows the information processing system 100 to consider whether or not it is necessary to acquire update information when generating the series of behavioral change content.

[0120] Furthermore, the information processing system 100 can evaluate the contribution of each of the multiple update pieces of information and determine whether or not to acquire each of the multiple update pieces of information based on their respective contributions. This allows the information processing system 100 to select, for example, only the update pieces of information that can be used to update a content generation model capable of generating more effective behavioral change content.

[0121] This embodiment is provided to facilitate understanding of the present invention and is not intended to limit its interpretation. The present invention may be modified or improved without departing from its spirit, and equivalents thereof are also included.

[0122] Furthermore, in the present invention, "part" does not merely mean a physical means, but also includes cases where the functions of that "part" are realized by software. Also, even if the functions of one "part" or device are realized by two or more physical means, devices, or software, the functions of two or more "parts" or devices may be realized by one physical means, device, or software.

Claims

1. An information processing system comprising: a contribution evaluation unit that evaluates the contribution of update information to updating a content generation model, which generates behavioral change content that causes a user to perform a predetermined target action, based on a contribution evaluation rule that evaluates the contribution of update information to updating the content generation model; an acquisition necessity determination unit that determines whether or not to acquire the update information based on the contribution; an update information acquisition unit that acquires the update information based on the result of the necessity determination; and a model update unit that updates the content generation model using the acquired update information.

2. The information processing system according to claim 1, further comprising: a behavioral characteristics acquisition unit that acquires behavioral characteristics information relating to the user's behavioral characteristics included in the update information, wherein the model update unit updates the content generation model using the behavioral characteristics information.

3. The information processing system according to claim 1 or 2, further comprising: a content evaluation unit that evaluates the effect of behavioral change content on causing a user to perform a predetermined target behavior, based on content evaluation rules for evaluating the effect of behavioral change content on the updated behavioral change content obtained from an updated content generation model updated using the update information, wherein the contribution evaluation unit evaluates the contribution to the evaluated effect.

4. The information processing system according to claim 3, wherein the content evaluation unit further evaluates the effects of the pre-update behavior change content obtained from the pre-update content generation model before it is updated using the update information, based on the content evaluation rules, and the contribution evaluation unit evaluates the contribution based on the effects of the pre-update behavior change content and the effects of the post-update behavior change content.

5. The information processing system according to claim 3, wherein the contribution evaluation unit further evaluates the contribution based on external environment information relating to the external environment related to the execution of the predetermined target action.

6. The information processing system according to claim 3, wherein the behavior change content includes a series of behavior change content, each of which corresponds to a plurality of behaviors performed up to the predetermined target behavior, and the content evaluation unit evaluates the effect of the series of updated behavior change content obtained from the updated content generation model.

7. The information processing system according to claim 1 or 2, wherein the update information includes a plurality of update information, the contribution evaluation unit evaluates the contribution of each of the plurality of update information to the content generation model when each of the plurality of update information is used for the update, the acquisition necessity determination unit determines whether each of the plurality of update information is necessary to acquire, and the update information acquisition unit acquires the update information that has been determined to be necessary to acquire from among the plurality of update information.

8. An information processing method comprising: evaluating the contribution of update information to a content generation model, which generates behavioral change content that causes a user to perform a predetermined target action, based on a contribution evaluation rule that evaluates the contribution of update information to the content generation model; determining whether or not to acquire the update information based on the contribution; acquiring the update information based on the result of the determination of whether or not to acquire it; and updating the content generation model using the acquired update information.

9. A program that causes a computer to perform the following actions: evaluate the contribution of update information to a content generation model, which generates behavioral change content that causes a user to perform a predetermined target action, based on a contribution evaluation rule that evaluates the contribution of update information to the content generation model; determine whether or not to acquire the update information based on the contribution; acquire the update information based on the result of the determination of whether or not to acquire it; and update the content generation model using the acquired update information.

Images