Information output program, information output method, and information processing device

Abstract

To enable incorporation of human action choices in conditional branching options.SOLUTION: An information output program disclosed herein causes a computer to perform steps of identifying a flow of people along a route with multiple options using an action selection model configured to indicate an action a person may choose regarding a measure, generating information indicating a predicted result of the measure based on the identified flow of people, and outputting the generated information on the predicted result of the measure to a display screen.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information output program, an information output method, and an information processing device. [Background technology]

[0002] One type of workflow is the policy flow graph, which diagrams the process of assigning policy target objects, such as people, to services to achieve the policy's objectives in various fields such as medical care, nursing care, and administration.

[0003] For example, the effectiveness of a measure can be predicted by counting the routes from the start node of the flow graph of the measure, through conditional branches with multiple options, to the end node, for each person targeted by the measure.

[0004] In such conditional branching, one aspect is the use of quantitative data such as test values ​​from a health checkup or categorical data such as gender for condition determination. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2024-1987 [Patent Document 2] Japanese Patent Application Publication No. 2019-197372 [Patent Document 3] US Patent Application Publication No. 2019 / 0180868 [Patent Document 4] US Patent Application Publication No. 2018 / 0039949 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the conditional branching in the flow graph of the above measures can only perform conditional judgments using quantitative or categorical data, making it difficult to incorporate actions that may change choices in measures that encourage people to change their behavior into the conditional branching options.

[0007] In one aspect, an object is to provide an information output program, an information output method, and an information processing device that can incorporate human behavior selection into options for conditional branching. [Means for solving the problem]

[0008] In one aspect, the information output program causes a computer to execute the following process: using a behavioral selection model that indicates which behavior a person will choose in response to a measure, identify the flow of people along a route with multiple options; based on the identified flow of people, generate information indicating the predicted results of the measure; and output the generated information on the predicted results of the measure to a display screen. [Effects of the Invention]

[0009] According to one embodiment, human action choices can be incorporated into conditional branching options. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram illustrating an example of the functional configuration of a server device. [Figure 2] FIG. 2 is a diagram illustrating a flow graph of a measure. [Figure 3] FIG. 3 is a diagram showing a specific example of a flow graph of a measure. [Figure 4] FIG. 4 is a diagram illustrating an example of a service that encourages behavioral change. [Figure 5] FIG. 5 is a diagram illustrating an example of a policy flow. [Figure 6] FIG. 6 is a schematic diagram showing an example of setting a behavior selection model. [Figure 7] FIG. 7 is a diagram illustrating an example of model information. [Figure 8]FIG. 8 is a schematic diagram illustrating one aspect of graph searching. [Figure 9] FIG. 9 is a schematic diagram illustrating an example of information output. [Figure 10] FIG. 10 is a flowchart showing the procedure of the information output process. [Figure 11] FIG. 11 is a schematic diagram (1) illustrating an application example of information output. [Figure 12] FIG. 12 is a schematic diagram (2) illustrating an application example of information output. [Figure 13] FIG. 13 is a schematic diagram (3) illustrating an application example of information output. [Figure 14] FIG. 14 is a schematic diagram illustrating an application example of the behavior selection model. [Figure 15] FIG. 15 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, examples for implementing the information output program, information output method, and information processing device according to the present disclosure will be described with reference to the accompanying drawings. Note that this example merely illustrates one example or aspect, and the structure, action, function, properties, characteristics, methods, uses, etc. according to the present disclosure are not limited by such examples.

[0012] Example 1 <System configuration> Fig. 1 is a block diagram showing an example of the functional configuration of the server device 10. Fig. 1 shows the server device 10 that provides an information output function that incorporates human behavior selection into options for conditional branching on a flow graph of a policy and outputs the predicted results of the effects of the policy.

[0013] The server device 10 can provide the above information output function as a cloud service by executing a platform as a service (PaaS) type middleware or a software as a service (SaaS) type application.

[0014] As shown in Fig. 1, a server device 10 can be communicatively connected to a client terminal 30 via a network NW. For example, the network NW may be any type of communication network, whether wired or wireless, such as the Internet or a local area network (LAN). Note that Fig. 1 shows an example in which one client terminal 30 is connected to one server device 10, but any number of client terminals 30 may be connected.

[0015] The client terminal 30 is a terminal device that receives the information output function. For example, the client terminal 30 may be used by a customer such as a policy planner, who is an example of an entity that implements a policy, such as a local government or other related party. The client terminal 30 may be realized by any computer, such as a personal computer, a smartphone, a tablet terminal, or a wearable terminal.

[0016] Although the above example illustrates the information output function being provided as a cloud service, the present invention is not limited to this. For example, the above information output function may be provided on-premise. Furthermore, the above example illustrates the information output function being provided in a client-server system, but the present invention is not limited to this. For example, the above information output function may be provided standalone by causing an application running on the client terminal 30 to cause the client terminal 30 to execute processing corresponding to the above information output function.

[0017] <Measure model> Here, the term "policy model" includes a diagram consisting of multiple components that define the content of a policy in a hierarchical structure. For example, a policy model can be realized as a policy flow graph, a social concept, an organizational chart, or medical guidelines.

[0018] The following explanation will be given using a policy flow graph as an example of one policy model. Figure 2 is a diagram illustrating a policy flow graph. Z1, Z2, Z3, and Z4 in Figure 2 represent, for example, services that an administrator provides to a user. These are sometimes referred to as "service implementation components." Specific examples of services include, in the medical field, "interventions" to which the policy target object, such as residents, is assigned, such as undergoing a health checkup or being examined by a specialist, as well as "no intervention" such as follow-up observation, but are not limited to policies in the medical field.

[0019] H1 and H2 indicate, for example, conditional branches including conditions. These may also be referred to as "conditional branch components." Specific examples of conditions, in the medical field, include an estimated glomerular filtration rate (eGFR) below a threshold, a hemoglobin A1c value (HbA1c) below a threshold, and a urinary protein value equal to or greater than a threshold, but are not limited to medical conditions.

[0020] Z1, Z2, Z3, Z4, H1, and H2 may each be referred to as a "component." From the perspective of graph data, such a "component" may correspond to an example of a "node." Furthermore, the connection between nodes may correspond to an example of an "edge," including a "directed edge."

[0021] In this embodiment, policy planning in the medical field will be described as an example, but the use case of the information output function described above is not limited to this. For example, the information output function described above may be applied to various policy planning in the transportation field such as road pricing, the energy field such as decarbonization and power supply, as well as work, testing, questionnaires, etc. In this case, the same effects as when applied to policy planning in the medical field can be obtained.

[0022] Figure 3 is a diagram showing a specific example of a policy flow graph. As shown in Figure 3, a policy is modeled as a workflow consisting of a combination of components such as conditional branching and service implementation. Then, the number of people receiving each service is output from a model that has been trained by accumulating information and parameters on the flow of people based on the actual values ​​when each conditional branching component is used.

[0023] In the example shown in FIG. 3, the number of people N=1000 is input at S0. At S1, component #1 as service execution component A is set to "health check." At S2, component #2 as conditional branch component B is set to "eGFR<α." If "eGFR<α" is not satisfied (see the NO route at S2), it is determined that there is "no intervention" by a specialist for the citizen, as shown at S5.

[0024] On the other hand, if "eGFR<α" is satisfied (see the YES route at symbol S2), then component #3 as conditional branch component C is set to "HbA1c<β" as shown at symbol S3. If "HbA1c<β" is satisfied (see the YES route at symbol S3), then component #4 as conditional branch component D is set to "nephrologist" as shown at symbol S6, and it is determined that the citizen requires intervention by a "nephrologist." On the other hand, if "HbA1c<β" is not satisfied (see the NO route at symbol S3), then it is determined that the citizen requires intervention by a "diabetes specialist" as shown at symbol S7.

[0025] In the example shown in Figure 3, the number of people flowing through part #1, part #2, part #3, and part #4 in that order is predicted, as indicated by the arrows. For example, in the flow graph of the measure shown in Figure 3, if the number of people N=1000 is assigned to interventions Z2 to Z4, the results will be as follows: 50 people are assigned to intervention Z2. 150 people are assigned to intervention Z3. Furthermore, 800 people are assigned to intervention Z4.

[0026] Hereinafter, the flow graph of a policy may be abbreviated as "policy flow." Policy flows may be shared in any framework. As just one example, policy flows can be shared among organizations around the world, such as public organizations such as local governments, through a data-based platform that allows policy flows to be shared, cross-referenced, and updated.

[0027] Such a data infrastructure platform may be provided by the business that provides the information output function described above, or by another business. For example, a policy planner can refer to the policy flows from around the world collected in the data infrastructure via a client terminal 30. In this case, by incorporating all or part of the policy flows collected in the data infrastructure, the existing policy flows can be updated to support the planning of original policy flows.

[0028] In addition to sharing policy flows, the data infrastructure can also provide the following back-end functions. For example, it can run simulations that mimic the flow of objects targeted by services on existing and proposed policy flows. It can also evaluate indicators of existing and proposed policy flows, such as effectiveness and cost, and compare existing and proposed policy flows, or compare policy flows across multiple proposals.

[0029] <One aspect of the issue> As explained in the Background Technology section above, the conditional branching in the policy flow related to the above-mentioned conventional technology can only perform conditional judgments using quantitative data or categorical data, making it difficult to incorporate behaviors that may change choices through policies that encourage people to change their behavior into the options for the conditional branching.

[0030] One example of where incorporating behavioral choices into the conditional branching options of a policy flow has technical value is the planning of health policies.

[0031] In order to extend the healthy life expectancy of its citizens, the government is requesting that local governments and companies formulate data health plans and implement efficient health programs.

[0032] In response to such requests, insurers such as companies are implementing health programs such as encouraging people to undergo health checkups and providing health guidance, and are planning and implementing various measures every year to achieve the targets of the data health plans they have formulated.

[0033] For example, in order to increase the rate of people undergoing health checkups, the company provides a variety of services, such as subsidizing health checkup costs, supporting health checkup reservations through a call center, and encouraging people to take part in health checkups using various channels such as postcards, emails, and SNS (Social Networking Services).

[0034] These services are being reviewed for improvement, and there is a high need to estimate in advance how much the attendance rate will increase if the service is changed, and to determine how changes should be made based on cost-effectiveness, etc.

[0035] Figure 4 is a diagram showing an example of a service that encourages behavioral change. As shown in Figure 4, the recommendation to undergo a health checkup encourages the recipient to take action to undergo a health checkup. Whether the person who receives this recommendation postcard actually chooses to undergo a health checkup or chooses not to do anything is left to the individual's discretion.

[0036] The technical value of predicting the effect of increasing the number of people taking health checkups on a policy flow that incorporates conditional branching that changes people's behavior by changing the message on postcards encouraging people to take checkups is high.

[0037] <One aspect of the problem-solving approach> Therefore, the information output function of this embodiment uses a behavioral selection model associated with the conditional branch of the policy flow to predict the route to be assigned by the options of the conditional branch, and outputs the flow of people on the policy flow, thereby incorporating people's behavioral selection into the options of the conditional branch.

[0038] <Configuration of Server Device 10> Next, the functional configuration of the server device 10 that provides the above-mentioned information output function will be described. Fig. 1 shows a block diagram related to the information output function of the server device 10. As shown in Fig. 1, the server device 10 has a communication control unit 11, a storage unit 13, and a control unit 15. Note that Fig. 1 only shows an excerpt of the functional units related to the above-mentioned information output function, and the server device 10 may also be provided with functional units other than those shown.

[0039] The communication control unit 11 is a functional unit that controls communication with other devices such as the client terminal 30. In one embodiment, the communication control unit 11 can be realized by a network interface card such as a LAN card. In one aspect, the communication control unit 11 receives a request from the client terminal 30 to output information related to the effectiveness of the measures, or outputs a predicted result of the effectiveness of the measures to the client terminal 30.

[0040] The storage unit 13 is a functional unit that stores various types of data. In one embodiment, the storage unit 13 may be realized by internal, external, or auxiliary storage of the server device 10. For example, the storage unit 13 stores a flow DB (DataBase) 13A, a model information DB 13B, and a personal information DB 13C. The flow DB 13A, the model information DB 13B, and the personal information DB 13C will be described later together with a scene in which reference or registration is performed.

[0041] The control unit 15 is a functional unit that performs overall control of the server device 10. For example, the control unit 15 can be realized by a hardware processor. As shown in FIG. 1, the control unit 15 has a registration unit 15A, a reception unit 15B, an identification unit 15C, a generation unit 15D, and an output unit 15E. The control unit 15 may also be realized by hardwired logic or the like.

[0042] The registration unit 15A is a processing unit that registers a policy flow. In one embodiment, the registration unit 15A newly registers a policy flow created by the client terminal 30 in the flow DB 13A stored in the storage unit 13, or overwrites and saves an edited policy flow that has already been registered in the flow DB 13A.

[0043] The collection of policy flows managed by the flow DB 13A may include templates of existing policies from around the world collected in the data base. For example, when formulating a policy, it is important to consider whether similar policies have been implemented in the past, from the perspective of administrative (political) ease of implementation. From this perspective, the original policy planner's original plan can be updated by incorporating all or part of existing policies similar to the original plan from the templates collected in the data base.

[0044] Here, when a policy flow is created in which human behavioral choices are incorporated into the options for conditional branching, a behavioral selection model is set up that associates services that encourage people to change their behavior in the policy flow with the behavioral options for conditional branching.

[0045] Figure 5 is a diagram showing an example of a policy flow. Figure 5 shows a policy flow F11 that was created by partially modifying an existing policy flow f1 that was registered in the flow DB 13A. As shown in Figure 5, the policy flow F11 includes, as examples of services that encourage people to change their behavior, a service S1 that sends a postcard A encouraging them to see a doctor and a service S2 that accepts appointments for health checkups.

[0046] Of these, in policy flow F11, a change has been made to the message of the postcard O encouraging a medical examination sent by service S1 in the existing policy flow f1. That is, in service S1 of policy flow F11, a change has been made to send a postcard A encouraging a medical examination, in which "You can receive a medical examination for free" has been added to the default message of postcard O encouraging a medical examination sent by service S1 in the existing policy flow f1.

[0047] Fig. 6 is a schematic diagram showing an example of the setting of a behavior selection model. Fig. 6 shows an example of a setting screen 200 in which the service S1 (medical examination recommendation postcard A) of the policy flow F11 shown in Fig. 5 is specified as a service that affects the behavior selection, and the conditional branch C1 of the policy flow F11 shown in Fig. 5 is specified as a conditional branch related to the behavior selection.

[0048] Here, the setting screen 200 shows an example of an algorithm for a behavioral selection model in which a simplified behavioral model is selected that generates a random number corresponding to the index of each option in a conditional branch according to a set probability that the option will be selected.

[0049] When such a simple behavioral model is used, the probability of selecting service S2 (reserving a health checkup) and the probability of selecting service S4 (not undergoing a health checkup) can be received on the setting screen 200. For example, in the example of the setting screen 200 shown in Fig. 6, a behavioral selection model that realizes an estimate of the expected effect that 80% of people who received service S1 will select service S2, while 20% of people who received service S1 will select service S4, is associated with service S1.

[0050] In this case, the registration unit 15A additionally registers the model information 13B1 input via the setting screen 200 in the model information DB 13B stored in the storage unit 13.

[0051] FIG. 7 is a diagram showing an example of model information 13B1. As shown in FIG. 7, model information 13B1 may be data in which a service name, a state quantity to be updated, an algorithm of a behavior selection model, and usage data of the behavior selection model are associated with each other. According to this model information 13B1, a behavior selection model is set that is associated with service S1 of action flow F11 and selects the behavior of conditional branch C1 among the state quantities of a person to which action flow F11 is applied. Furthermore, a simplified behavior model is set in the algorithm of the behavior selection model. Furthermore, the probability distribution used by the simplified behavior model to generate random numbers is set to an 80% probability of selecting service S2 (reserving a health check) and a 20% probability of selecting service S4 (not having undergone a health check).

[0052] Note that Figures 6 and 7 show an example in which a simple behavioral model is used as an algorithm for the behavioral selection model, but this does not prevent the use of other algorithms such as an attribute information-based behavioral model that uses attribute information for behavioral selection or a discrete choice model.

[0053] Returning to the explanation of FIG. 1, the reception unit 15B is a processing unit that receives various information from the client terminal 30. In one embodiment, the reception unit 15B can receive a request from the client terminal 30 to output information regarding the effects of a policy. When receiving such an information output request, the reception unit 15B can receive the designation of one or more policy flows. For example, from the perspective of comparing the effects of multiple policies, it is possible to designate an existing policy flow and a policy flow of a draft to be planned, or to designate policy flows of multiple drafts.

[0054] The identification unit 15C is a processing unit that identifies the flow of a person along a route having multiple options using a behavior selection model that indicates which behavior a person will select in response to a policy. In one aspect, the identification unit 15C executes the following process for each of the M policy flows specified in the information output request received by the reception unit 15B. Specifically, the identification unit 15C inputs personal information, such as age, gender, and health checkup results, of a target group of the policy, e.g., personal information of insured persons enrolled in a health insurance plan, from the personal information DB 13C stored in the storage unit 13, into the m-th policy flow. As a result, the identification unit 15C identifies a route for each individual included in the target group of the policy, from the start node of the policy flow via a conditional branch node to the end node.

[0055] More specifically, for each of the N individuals included in the target group of the policy, the identification unit 15C performs a graph search to search for nodes by tracing edges while prioritizing the depth direction of the policy flow from the start node to the end node. That is, the identification unit 15C transitions the nth person to the next node. At this time, if model information including a service name corresponding to the node after the transition in a data entry exists in the model information DB 13B, the identification unit 15C generates a behavior selection model according to the settings defined in the model information. Then, the identification unit 15C uses the behavior selection model to select one behavior option from the behavior options for the next conditional branch. Then, the identification unit 15C performs an update to register the behavior option selected by the behavior selection model as the behavior for the next conditional branch in the state quantity of the nth person. Furthermore, if the node after the transition is a conditional branch, the identification unit 15C sets the route to an edge corresponding to the behavior option stored as the behavior of the conditional branch among the state quantities of the nth person. This graph search is repeated until the end node is reached.

[0056] Fig. 8 is a schematic diagram illustrating one aspect of graph search. Fig. 8 illustrates a graph search for a person identified by personal ID "1," and also shows excerpts of the state quantities for personal ID "1" at times t0, t1, and t2.

[0057] As shown in Fig. 8, time t0 corresponds to the start time. At time t0, the person with personal ID "1" is located at the start node, which is the initial position.

[0058] Next, at time t1, the person with personal ID "1" transitions to service S1 (medical examination recommendation postcard A). At time t1, a behavioral option for the next conditional branch C1 is selected using a behavior selection model generated according to the settings defined in model information 13B1 shown in FIG. 7. That is, in the example of model information 13B1 shown in FIG. 7, the behavioral selection model indicates that 80% of the people who received service S1 select service S2, while 20% of the people who received service S1 select service S4. Here, FIG. 8 shows an example in which the behavioral selection model selects service S2 (medical examination appointment) as a behavioral option for the next conditional branch C1. In this case, as shown by hatching in FIG. 8, the identification unit 15C executes an update to register service S2 (medical examination appointment) selected by the behavior selection model in the state quantity of the person with personal ID "1" as the behavior for the next conditional branch C1.

[0059] Then, at time t2, the person with personal ID "1" transitions to conditional branch C1. At this time t2, an edge corresponding to service S2 (medical checkup appointment), which is held as an action of conditional branch C1 among the state quantities of the person with personal ID "1", is set as the path. As a result, at the next time t3, the person with personal ID "1" can transition to service S2 (medical checkup appointment).

[0060] After that, the graph search is performed in the same manner from time t3 onwards until the terminal node is reached. As a result of this graph search, a route that passes through the nodes of service S1, conditional branch C1, service S2, conditional branch C2 and service S3 is obtained as the route to be assigned to person with individual ID "1" by policy flow F11.

[0061] The generation unit 15D is a processing unit that generates information indicating the predicted results of the effects of measures based on the flow of people identified by the identification unit 15C. In one embodiment, the generation unit 15D executes the following process for each of the M measure flows specified in the information output request received by the reception unit 15B. That is, the generation unit 15D counts the number of people passing through each edge included in the m-th measure flow based on the route on the m-th measure flow assigned to each individual by the identification unit 15C. The generation unit 15D then sets the display form of the edge included in the m-th measure flow, such as its thickness, depending on the number of people passing through the edge. For example, the generation unit 15D can set a thicker edge as the number of people passing through increases, and a thinner edge as the number of people passing through decreases. In this way, the generation unit 15D generates display data for each of the M measure flows, in which the thickness of each edge is set according to the number of people passing through the edge.

[0062] The output unit 15E executes output control for the client terminal 30. In one aspect, the output unit 15E causes the client terminal 30 to display the display data of the M number of policy flows generated by the generation unit 15D as a response to the information output request received by the reception unit 15B.

[0063] Fig. 9 is a schematic diagram for explaining an example of information output. Fig. 9 shows examples of display of a policy flow f1 including a service S1 that sends a postcard O encouraging a medical examination with a default message, and a policy flow F11 that includes a service S1 that sends a postcard A encouraging a medical examination with a message that has been changed to add "You can receive a medical examination for free" to the default message.

[0064] As shown in Figure 9, in policy flow f1, 60% of people who received service S1 select service S2, while 40% of people who received service S1 select service S4. On the other hand, in policy flow F11, 80% of people who received service S1 select service S2, while 20% of people who received service S1 select service S4.

[0065] It is possible to estimate the extent to which such changes in behavioral choices can be expected to have an effect on improving the participation rate in health checkups, which is the policy's goal. For example, in the example shown in Figure 9, the number of people "45" is displayed in association with the edge connecting conditional branch C2 and service S3 in policy flow f1, while the number of people "60" is displayed in association with the edge connecting conditional branch C2 and service S3 in policy flow F11. This makes it possible to estimate the effect of changing postcard O encouraging health checkups to postcard A encouraging health checkups, which will increase the number of people who take health checkups by 15, i.e., increase the participation rate in health checkups by 1.3 times.

[0066] <Processing flow> Next, the flow of processing of the server device 10 according to this embodiment will be described. Fig. 10 is a flowchart showing the procedure of the information output processing. This processing can be started when an information output request is accepted by the accepting unit 15B, as an example. Note that the start condition of this processing is not limited to the acceptance of an information output request, but can also be when a new action flow is registered or when an existing action flow is overwritten and saved.

[0067] 10, when an information output request is received (step S101), the identification unit 15C executes loop processing 1, which repeats the processing from step S102 to step S109 described below a number of times corresponding to the number M of policy flows specified in the information output request received in step S101. Note that the processing from step S102 to step S109 described below can also be executed in parallel for each of the M policy flows.

[0068] Furthermore, the identification unit 15C executes loop processing 2, which repeats the processing from step S102 to step S109 described below, the number of times corresponding to the total number N of people to be applied to the graph search of the m-th policy flow. Note that the processing from step S102 to step S109 described below can also be executed in parallel for each of the M policy flows.

[0069] Furthermore, the identification unit 15C executes loop processing 2, which repeats the processing from step S102 to step S108 below until the graph search of the nth person reaches the terminal node of the mth policy flow. Note that the processing from step S102 to step S108 below can also be executed in parallel for each of the M policy flows.

[0070] That is, the identification unit 15C transitions the n-th person to the next node (step S102). At this time, if model information including a service name corresponding to the node after the transition in a data entry exists in the model information DB 13B (step S103: Yes), the identification unit 15C executes the following process.

[0071] For example, the identification unit 15C selects one action option from the action options for the next conditional branch using the action selection model generated according to the settings defined in the model information (step S104). Then, the identification unit 15C executes an update to register the action option selected in step S104 as an action for the next conditional branch in the state quantity of the n-th person (step S105).

[0072] Furthermore, if the node after the transition is a conditional branch (No in step S103 and Yes in step S106), the specification unit 15C refers to the state quantity of the n-th person (step S107). Then, the specification unit 15C sets an edge corresponding to the action option held as the action of the conditional branch among the state quantities of the n-th person as the course (step S108).

[0073] By repeating this loop process 3, it is possible to search for a route to be assigned to the nth person by the mth action flow.

[0074] Thereafter, the generating unit 15D increments the number of passers of the edges that overlap with the route allocated to the nth person by the mth policy flow, among the edges included in the mth policy flow (step S109).

[0075] By repeating this loop process 2, the routes of N people in the m-th policy flow and the number of people passing through each edge are obtained, thereby generating display data for the m-th policy flow.

[0076] Furthermore, by repeating the loop process 1, display data is generated for each of the M action flows.

[0077] Thereafter, the output unit 15E causes the client terminal 30 to display the display data of the M number of action flows generated as a result of the loop processing 1 to the loop processing 3 (step S110), and ends the processing.

[0078] <One aspect of the effect> As described above, the server device 10 according to the present embodiment uses a behavior selection model associated with a conditional branch of a policy flow to predict routes allocated by options of the conditional branch, and outputs the flow of people on the policy flow. Therefore, the server device 10 according to the present embodiment can incorporate people's behavior selection into options of a conditional branch.

[0079] <Example 2> Although the embodiments of the present disclosure have been described above, various applications are possible, and the present disclosure may be implemented in various different forms other than the above-described embodiments.

[0080] <Exercise creative ability> The matters explained in the first embodiment, such as the specific examples of the policy flow to be output and the types of algorithms of the behavior selection model, are merely examples and can be changed. Also, the order of processing in the flowchart explained in the embodiment can be changed within a consistent range.

[0081] <Information output application example 1> In the above Example 1, as examples of modifications to service S1, action flow f1 was given in which a postcard O encouraging a patient to undergo a medical examination is sent containing a default message, and action flow F11 in which a postcard A encouraging a patient to undergo a medical examination is sent with a message modified to add "You can receive a medical examination free of charge" to the default message. However, action flows in which other postcard encouraging a patient to undergo a medical examination with other modified messages may also be displayed.

[0082] Figure 11 is a schematic diagram (1) that explains an application example of information output. For the sake of convenience, Figure 11 condenses the display of policy flow f1, which has already been explained using Figure 9. As shown in Figure 11, in addition to policy flow f1 and policy flow F11 shown in Figure 9, policy flow F12 is displayed, which includes service S1 that sends postcards B encouraging medical checkups with a message that has been modified to add "Anyone can get cancer, so get a health checkup" to the default message.

[0083] In this policy flow F12, 70% of the people who received service S1 select service S2, while 30% of the people who received service S1 select service S4.

[0084] By displaying this type of action flow F12, the action flow f1, action flow F11, and action flow F12 can be compared as follows. For example, the number of people "45" is displayed in association with the edge connecting the conditional branch C2 and service S3 of action flow f1. On the other hand, the number of people "60" is displayed in association with the edge connecting the conditional branch C2 and service S3 of action flow F11, and the number of people "52" is displayed in association with the edge connecting the conditional branch C2 and service S3 of action flow F11. Therefore, it is possible to evaluate whether the proposed message changes for both the health checkup recommendation postcard A and the health checkup recommendation postcard B are more effective in increasing the health checkup attendance rate than the default message. Furthermore, it can be determined that the health checkup recommendation postcard A is more effective in increasing the health checkup attendance rate than the health checkup recommendation postcard B.

[0085] <Information output application example 2> In the first embodiment, an example in which the service S1 is changed is given, but the change is not limited to the change of the service S1, and similar information output can be realized even when other services are changed.

[0086] Fig. 12 is a schematic diagram (2) illustrating an application example of information output. Fig. 12 shows an example of displaying a policy flow f1 including a service S2 that uses the current reservation system O as the reservation system for health checkups, and a policy flow F21 including a service S2 in which the reservation system for health checkups has been changed to reservation system A.

[0087] As shown in Figure 12, in policy flow f1, 75% of people who received service S2 select service S3, while 25% of people who received service S2 select service S4. On the other hand, in policy flow F21, 90% of people who received service S2 select service S3, while 10% of people who received service S2 select service S4.

[0088] It is possible to estimate the extent to which such a change in behavioral choice can be expected to have an effect on improving the participation rate in health checkups, which is the policy goal. For example, in the example shown in Figure 12, the number of people "45" is displayed in association with the edge connecting conditional branch C2 and service S3 in policy flow f1, while the number of people "54" is displayed in association with the edge connecting conditional branch C2 and service S3 in policy flow F21. This makes it possible to estimate the effect of changing the health checkup reservation system from reservation system O to reservation system A, which will increase the number of people who undergo health checkups by 9, i.e., increase the participation rate in health checkups by 1.2 times.

[0089] <Information output application example 3> In the above application example 1, the action flow F12 and the action flow F21 are displayed as two change proposals for the same type of service S1, but it is also possible to display change proposals for different services.

[0090] Figure 13 is a schematic diagram (3) that explains an application example of information output. Figure 13 shows an example display of a policy flow F12 that includes service S1, which sends postcards B encouraging people to get checked out, with the message "Anyone can get cancer, so get a health checkup" added to the default message, and a policy flow F21 that includes service S2, in which the health checkup reservation system has been changed to reservation system A. Furthermore, Figure 13 displays the cost of changing service S1, "100,000 yen," in association with policy flow F12, and the cost of changing service S2, "1,000,000 yen," in association with policy flow F21.

[0091] For example, in the example shown in Figure 13, the number of people "52" is displayed in association with the edge connecting conditional branch C2 of policy flow F11 and service S3, and the number of people "54" is displayed in association with the edge connecting conditional branch C2 of policy flow F21 and service S3. This visualizes that the number of people who underwent health checkups (participation rate) is higher in policy flow F21, which is a proposed change to service S2, than in policy flow F12, which is a proposed change to service S1. Furthermore, by visualizing the cost of the proposed change to service S1 (100,000 yen) and the cost of the proposed change to service S2 (1,000,000 yen), cost-effectiveness, or so-called cost performance, can be presented. For example, when selecting a policy with cost as the priority, policy flow F12, which is a proposed change to service S1, can be adopted.

[0092] <Application example of behavioral choice model> In the above-mentioned first embodiment, an example was given in which a simplified behavior model is used as an algorithm for a behavior selection model, but an attribute information-based behavior model in which attribute information is used for behavior selection can also be used.

[0093] For example, in the simple behavioral model, each individual has the same probability of selecting an action, but the attribute information-based behavioral model can change the probability of selecting a conditional branching action depending on personal attribute information, such as age.

[0094] Such personal attribute information can be obtained as statistical data through experiments, for example, sampling surveys. Figure 14 is a schematic diagram explaining an application example of a behavioral choice model. Figure 14 shows an example of building a behavioral choice model using data obtained from an experiment to see whether the behavior of booking a health checkup changes in response to two proposed message changes, namely, a health checkup recommendation postcard A and a health checkup recommendation postcard B, such as personal attributes such as age, as explanatory variables.

[0095] 14 shows experimental data 20A obtained when a recommendation to have a medical checkup was made using message change plan A, which adds "You can have a medical checkup for free" to the default message, and experimental data 20B obtained when a recommendation to have a medical checkup was made using message change plan B, which adds "Anyone can get cancer, so get a medical checkup" to the default message. Using experimental data 20A or experimental data 20B, a logistic regression analysis was performed using age as the explanatory variable and whether or not to make an appointment for a medical checkup as the objective variable.

[0096] This allows a behavioral selection model 21A to be generated from the experimental data 20A, and a behavioral selection model 21B to be generated from the experimental data 20B. For example, in the case of the experimental data 20A, a behavioral selection model 21A is generated that selects the behavior of making an appointment for a health check with a uniform probability across all ages. On the other hand, in the case of the experimental data 20B, a behavioral selection model 21B is generated that increases the probability of selecting the behavior of making an appointment for a health check as the age decreases.

[0097] Here, we have given an example of using logistic regression as an example of an attribute information usage behavior model, but in the case of a discrete choice model, an attribute information usage behavior model can be realized by formulating deterministic terms corresponding to individual attributes.

[0098] <Digital Twin> The server device 10 can also perform a simulation of the flow of objects that are the target of the service on a digital twin that virtually reproduces a real space. First, the server device 10 generates a digital twin that reproduces a real space in a virtual space. Next, the server device 10 uses a behavioral selection model to simulate the flow of people along a route with multiple options in the generated digital twin. Then, the server device 10 generates information indicating the predicted results of the measures based on the results of the performed simulation.

[0099] More specifically, the server device 10 generates, for example, a digital twin in a virtual space that is time-synchronized with the real space. For example, the server device 10 collects data related to personal attribute information of a person from a terminal and updates the personal attribute information of the person in real time in the digital twin. Then, for example, the server device 10 identifies a group of agents corresponding to the currently existing person in the digital twin. Also, for example, the server device 10 inputs the identified group of agents into a policy flow in the digital twin. Then, the server device 10 generates information indicating the predicted results of the policy based on the input results of the policy flow.

[0100] More specifically, for each individual included in a group of agents, the server device 10 identifies a route by which the individual is allocated from the start node of the policy flow via a conditional branch node to the end node. For example, the server device 10 performs a simulation on the digital twin to determine whether each of the multiple individuals will take action in response to a policy using attribute information of each of the multiple agents corresponding to each of the multiple individuals and a behavior selection model associated with a conditional branch of the policy model. Then, based on the results of the simulation, the server device 10 identifies a route to which each of the multiple individuals will be allocated based on the options of the conditional branch of the policy model. Then, the server device 10 generates information indicating a predicted result of the policy, including a route to which each of the identified multiple individuals will be allocated. This allows the server device 10 to perform a simulation with high reproducibility. Furthermore, the server device 10 can improve the accuracy of the predicted result of the policy.

[0101] <System> The information including the processing procedures, control procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, any one or more of the functional units of the registration unit 15A, the reception unit 15B, the identification unit 15C, the generation unit 15D, and the output unit 15E of the server device 10 may be configured as separate devices.

[0102] Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown. In other words, all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Note that each configuration may also be a physical configuration.

[0103] Furthermore, each processing function performed by each device can be realized, in whole or in part, by a CPU (Central Processing Unit) and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[0104] <Hardware> Next, an example of the hardware configuration of the computer described in the above embodiment will be described. Fig. 15 is a diagram showing an example of the hardware configuration. As shown in Fig. 15, the server device 10 has a communication device 10a, a storage device 10b, a memory 10c, and a processor 10d. Note that the components shown in Fig. 15 may be connected to each other via a bus or the like.

[0105] The communication device 10a is a network interface card, etc. The storage device 10b is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD). For example, the storage device 10b stores programs and databases that operate the functions shown in FIG.

[0106] The processor 10d reads out a program that executes the same processing as the processing unit shown in FIG. 1 from the storage device 10b or the like and loads it into the memory 10c, thereby operating a process that executes the functions described in FIG.

[0107] Such a process realizes the same functions as the processing units of the information processing device 10. For example, the processor 10d reads out a program having the same functions as the registration unit 15A, the reception unit 15B, the identification unit 15C, the generation unit 15D, the output unit 15E, etc. from the storage device 10b, etc. Then, the processor 10d executes a process that executes the same processes as the registration unit 15A, the reception unit 15B, the identification unit 15C, the generation unit 15D, the output unit 15E, etc.

[0108] In this way, the information processing device 10 operates as an information processing device that executes an information output method by reading and executing a program. The information processing device 10 can also realize functions similar to those of the above-described embodiment by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program in these other embodiments is not limited to being executed by the information processing device 10. For example, the present invention can also be applied in the same way to cases where another computer or server executes the program, or where these execute the program in cooperation with each other.

[0109] The above program can be distributed via a network such as the Internet. The above program can also be recorded on any recording medium and executed by a computer by reading it from the recording medium. For example, the recording medium can be a hard disk, a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), a digital versatile disk (DVD), or the like.

[0110] The following additional notes are provided regarding the embodiments including the above examples.

[0111] (Appendix 1) Using a behavioral choice model that shows which behavior a person will choose in response to a policy, we identify the flow of people along a route with multiple options, generating information indicating a predicted result of the measure based on the identified flow of people; Output the generated forecast results information on the display screen. An information output program that causes a computer to execute a process.

[0112] (Supplementary Note 2) The computer further executes a process of acquiring data about the person, the identifying process includes a process of identifying a route to a terminal node via nodes and edges using a behavior selection model that indicates which behavior the person will select in response to the policy when the acquired data on the person is input into the policy model; the generating process includes a process of generating information indicating a predicted result of a measure regarding a service to be provided to the person based on the identified route. 2. The information output program according to claim 1,

[0113] (Supplementary Note 3) The process of specifying includes a process of predicting the number of people who will use the service on each of a plurality of routes from an intermediate node to a terminal node branched from the intermediate node, based on a behavior selection model that indicates which behavior a person will select in response to the measure; The generating process includes a process of generating information indicating a predicted result of a measure in which the predicted number of people is associated with the service to be provided to the person. 2. The information output program according to claim 1,

[0114] (Supplementary Note 4) The computer further executes a process of acquiring a policy model having a flow structure composed of nodes and edges, in which an action selection model indicating which action a person will select in response to the policy is associated with an intermediate node of the flow structure, and a service to be provided to the person is associated with an intermediate node or a terminal node of the flow structure; the specifying process includes a process of specifying a route to a terminal node by tracing the nodes and edges of the flow structure based on the behavior selection model when data on the person to be analyzed is input into the acquired policy model; the generating process includes a process of generating information indicating a predicted result of a measure, in which the number of people traveling along the identified route is associated with the service to be provided to the person; the outputting process includes a process of outputting information indicating the generated prediction result of the measure to the display screen. 2. The information output program according to claim 1,

[0115] (Supplementary Note 5) The process of identifying includes, when the acquired data on the person is input into the policy model, identifying the number of people flowing through the nodes of the flow graph based on the behavior selection model; The generating process generates an image with thickened edges based on the identified number of people, the outputting process displays the generated image on the display screen. 2. The information output program according to claim 1,

[0116] (Appendix 6) The generating process is a process of generating an image showing a first prediction result in which the number of people flowing through the nodes of the flow graph is predicted by inputting data about people into a first policy model before the policy is executed; and generating an image showing a second prediction result that predicts the number of people flowing through the nodes of the flow graph after the implementation of the policy by inputting the result into a second policy model after the policy is implemented, the outputting process includes a process of displaying an image indicating the first prediction result and an image indicating the second prediction result on the display screen. 2. The information output program according to claim 1,

[0117] (Appendix 7) Identify the behavioral choice model corresponding to the policy under consideration, storing the identified behavior selection model in association with an intermediate node of a second policy model after the policy is executed; 7. The information output program according to claim 6, further causing the computer to execute processing.

[0118] (Appendix 8) Creating a digital twin that recreates the real world in a virtual space, In the generated digital twin, using the behavioral selection model, a simulation of the flow of people along the route having the plurality of options is performed; generating information indicating a predicted result of the measures based on the results of the implemented simulation; 2. The information output program according to claim 1, further causing the computer to execute a process.

[0119] (Appendix 9) The process of generating the digital twin generates a digital twin in the virtual space that is time-synchronized with the real space, The process of performing the simulation uses attribute information of each of a plurality of agents corresponding to each of a plurality of people and the behavior selection model associated with a conditional branch of a policy model to perform a simulation on the digital twin as to whether each of the plurality of people will act on the policy, thereby identifying routes to which each of the plurality of people will be assigned based on options for the conditional branch; the generating process generates information indicating a predicted result of the measure, including a route to be allocated to each of the identified plurality of persons. 9. The information output program according to claim 8,

[0120] (Appendix 10) Using a behavioral choice model that shows which behavior a person will choose in response to a measure, the flow of people along a route with multiple options is identified, generating information indicating a predicted result of the measure based on the identified flow of people; Output the generated forecast results information on the display screen. An information output method characterized in that the processing is executed by a computer.

[0121] (Supplementary Note 11) The computer further executes a process of acquiring data about the person, the identifying process includes a process of identifying a route to a terminal node via nodes and edges using a behavior selection model that indicates which behavior the person will select in response to the policy when the acquired data on the person is input into the policy model; the generating process includes a process of generating information indicating a predicted result of a measure regarding a service to be provided to the person based on the identified route. 11. The information output method according to claim 10,

[0122] (Supplementary Note 12) The process of specifying includes a process of predicting the number of people who will use the service on each of a plurality of routes from an intermediate node to a terminal node branched from the intermediate node, based on a behavior selection model that indicates which behavior a person will select in response to the measure; The generating process includes a process of generating information indicating a predicted result of a measure in which the predicted number of people is associated with the service to be provided to the person. 11. The information output method according to claim 10,

[0123] (Supplementary Note 13) The computer further executes a process of acquiring a policy model having a flow structure composed of nodes and edges, in which an action selection model indicating which action a person will select in response to the policy is associated with an intermediate node of the flow structure, and a service to be provided to the person is associated with an intermediate node or a terminal node of the flow structure; the specifying process includes a process of specifying a route to a terminal node by tracing the nodes and edges of the flow structure based on the behavior selection model when data on the person to be analyzed is input into the acquired policy model; the generating process includes a process of generating information indicating a predicted result of a measure, in which the number of people traveling along the identified route is associated with the service to be provided to the person; the outputting process includes a process of outputting information indicating the generated prediction result of the measure to the display screen. 11. The information output method according to claim 10,

[0124] (Supplementary Note 14) The process of identifying includes, when the acquired data on the person is input into the policy model, identifying the number of people flowing through the nodes of the flow graph based on the behavior selection model; The generating process generates an image with thickened edges based on the identified number of people, the outputting process displays the generated image on the display screen. 11. The information output method according to claim 10,

[0125] (Appendix 15) The generating process is a process of generating an image showing a first prediction result in which the number of people flowing through the nodes of the flow graph is predicted by inputting data about people into a first policy model before the policy is executed; and generating an image showing a second prediction result that predicts the number of people flowing through the nodes of the flow graph after the implementation of the policy by inputting the result into a second policy model after the policy is implemented, the outputting process includes a process of displaying an image indicating the first prediction result and an image indicating the second prediction result on the display screen. 11. The information output method according to claim 10,

[0126] (Appendix 16) Identify the behavioral choice model corresponding to the policy under consideration, storing the identified behavior selection model in association with an intermediate node of a second policy model after the policy is executed; 16. The information output method according to claim 15, further comprising causing the computer to execute processing.

[0127] (Appendix 17) Using a behavioral choice model that shows which behavior a person will choose in response to a policy, we identify the flow of people along a route with multiple options, generating information indicating a predicted result of the measure based on the identified flow of people; Output the generated forecast results information on the display screen. An information processing device comprising a control unit that executes processing.

[0128] (Supplementary Note 18) The control unit further executes a process of acquiring data related to the person, the identifying process includes a process of identifying a route to a terminal node via nodes and edges using a behavior selection model that indicates which behavior the person will select in response to the policy when the acquired data on the person is input into the policy model; the generating process includes a process of generating information indicating a predicted result of a measure regarding a service to be provided to the person based on the identified route. 18. The information processing device according to claim 17.

[0129] (Supplementary Note 19) The process of specifying includes a process of predicting the number of people who will use the service on each of a plurality of routes from an intermediate node to a terminal node branched from the intermediate node, based on a behavior selection model that indicates which behavior a person will select in response to the measure; The generating process includes a process of generating information indicating a predicted result of a measure in which the predicted number of people is associated with the service to be provided to the person. 18. The information processing device according to claim 17,

[0130] (Supplementary Note 20) The control unit further executes a process of acquiring a policy model having a flow structure composed of nodes and edges, in which an action selection model indicating which action a person will select in response to the policy is associated with an intermediate node of the flow structure, and a service to be provided to the person is associated with an intermediate node or a terminal node of the flow structure; the specifying process includes a process of specifying a route to a terminal node by tracing the nodes and edges of the flow structure based on the behavior selection model when data on the person to be analyzed is input into the acquired policy model; the generating process includes a process of generating information indicating a predicted result of a measure, in which the number of people traveling along the identified route is associated with the service to be provided to the person; the outputting process includes a process of outputting information indicating the generated prediction result of the measure to the display screen. 18. The information processing device according to claim 17, [Explanation of symbols]

[0131] 10 Server device 11 Communication control section 13 Storage section 13A Flow DB 13B Model Information DB 13C Personal information DB 15 Control Unit 15A Registration Department 15B Reception 15C Specific part 15D generator 15E Output section 30 client terminals

Claims

1. Using a behavioral choice model that indicates which behavior a person will choose in response to a measure, the flow of people along a route with multiple options is identified; generating information indicating a predicted result of the measure based on the identified flow of the person; outputting the generated information on the predicted results of the measures to a display screen; An information output program that causes a computer to execute a process.

2. causing the computer to further perform a process of acquiring data about the person; the identifying process includes a process of identifying a route to a terminal node via nodes and edges using a behavior selection model that indicates which behavior the person will select in response to the policy when the acquired data on the person is input into the policy model; the generating process includes a process of generating information indicating a predicted result of a measure regarding a service to be provided to the person based on the identified route.

2. The information output program according to claim 1, wherein:

3. the specifying process includes a process of predicting the number of people who will use the service along each of a plurality of routes that start from an intermediate node and branch off to an end node, based on a behavior selection model that indicates which behavior a person will select in response to the measure; The generating process includes a process of generating information indicating a predicted result of a measure in which the predicted number of people is associated with the service to be provided to the person.

2. The information output program according to claim 1, wherein:

4. a process of acquiring a policy model having a flow structure composed of nodes and edges, in which an action selection model indicating which action a person will select in response to the policy is associated with an intermediate node of the flow structure, and in which a service to be provided to the person is associated with an intermediate node or a terminal node of the flow structure; the specifying process includes a process of specifying a route to a terminal node by tracing the nodes and edges of the flow structure based on the behavior selection model when data on the person to be analyzed is input into the acquired policy model; the generating process includes a process of generating information indicating a predicted result of a measure, in which the number of people traveling along the identified route is associated with the service to be provided to the person; the outputting process includes a process of outputting information indicating the generated prediction result of the measure to the display screen.

2. The information output program according to claim 1, wherein:

5. the identifying process includes, when the acquired data on the person is input into the policy model, identifying the number of people flowing through the nodes of the flow graph based on the behavior selection model; The generating process generates an image with thickened edges based on the identified number of people, the outputting process displays the generated image on the display screen.

2. The information output program according to claim 1, wherein:

6. The generating process includes: a process of generating an image showing a first prediction result in which the number of people flowing through the nodes of the flow graph is predicted by inputting data about people into a first policy model before the policy is executed; and generating an image showing a second prediction result that predicts the number of people flowing through the nodes of the flow graph after the implementation of the policy by inputting the result into a second policy model after the policy is implemented, the outputting process includes a process of displaying an image indicating the first prediction result and an image indicating the second prediction result on the display screen.

6. The information output program according to claim 1, wherein:

7. Identifying the behavioral choice model corresponding to the policy under consideration; storing the identified behavior selection model in association with an intermediate node of a second policy model after the policy is executed; 7. The information output program according to claim 6, further causing the computer to execute a process.

8. We create a digital twin that recreates the real world in a virtual space. In the generated digital twin, using the behavioral selection model, a simulation of the flow of people along the route having the plurality of options is performed; generating information indicating a predicted result of the measures based on the results of the implemented simulation; 2. The information output program according to claim 1, further causing the computer to execute a process.

9. The process of generating the digital twin generates a digital twin in the virtual space that is time-synchronized with the real space, The process of performing the simulation uses attribute information of each of a plurality of agents corresponding to each of a plurality of people and the behavior selection model associated with a conditional branch of a policy model to perform a simulation on the digital twin as to whether each of the plurality of people will act on the policy, thereby identifying routes to which each of the plurality of people will be assigned based on options for the conditional branch; the generating process generates information indicating a predicted result of the measure, including a route to be allocated to each of the identified plurality of persons.

9. The information output program according to claim 8, wherein:

10. Using a behavioral choice model that indicates which behavior a person will choose in response to a measure, the flow of people along a route with multiple options is identified; generating information indicating a predicted result of the measure based on the identified flow of the person; outputting the generated information on the predicted results of the measures to a display screen; An information output method characterized in that the processing is executed by a computer.

11. Using a behavioral choice model that indicates which behavior a person will choose in response to a measure, the flow of people along a route with multiple options is identified; generating information indicating a predicted result of the measure based on the identified flow of the person; outputting the generated information on the predicted results of the measures to a display screen; An information processing device comprising a control unit that executes processing.

Images