Measure output method, information processing device, and measure output program

Abstract

In this measure output method, a computer performs a process for: acquiring regional characteristics of a first region to which a measure is to be applied; referring to a storage unit that stores other measures applied to each of a plurality of second regions and regional characteristics of each of the plurality of second regions; calculating the degree of similarity between the regional characteristics of the first region and the regional characteristics of each of the plurality of second regions; and outputting a measure that is among the plurality of other measures stored in the storage unit, and that was applied to a similar region for which the degree of similarity is equal to or greater than a threshold value.

Description

Measure output method, information processing device, and measure output program

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

[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] 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, so it is becoming increasingly important to compare the flow graph of the proposed policy with the flow graph of existing policies that are used as reference.

[0004] From this perspective, it is common to extract similar measures that have a graph structure similar to that of the flow graph of the proposed measure from the flow graphs of existing measures to be used as reference.

[0005] Japanese Patent Application Laid-Open No. 2021-89523

[0006] However, just because an existing policy has a similar graph structure to the flow graph of the proposed policy does not necessarily mean that it is a similar policy. For example, if there are discrepancies in various characteristics, such as population distribution and demographics, between the region where the proposed policy is implemented and the region where the existing policy is implemented, even if the existing policy has a similar graph structure to the proposed policy, it may not be a similar policy to the proposed policy.

[0007] In one aspect, the present invention aims to provide a measure output method, an information processing device, and a measure output program that can output measures similar to an original plan.

[0008] In one embodiment of the policy output method, a computer acquires regional characteristics of a first region to which a policy is to be applied, references a memory unit that stores other policies applied to each of multiple second regions and the regional characteristics of each of the multiple second regions, calculates the similarity between the regional characteristics of the first region and each of the multiple second regions, and outputs policies applied to similar regions from among the multiple other policies stored in the memory unit whose similarity is equal to or greater than a threshold.

[0009] According to one embodiment, measures similar to the original plan can be output.

[0010] FIG. 1 is a block diagram showing an example of the functional configuration of a server device. FIG. 2 is a diagram illustrating an example of a flow graph of a policy. FIG. 3 is a diagram illustrating a specific example of a flow graph of a policy. FIG. 4 is a diagram illustrating an example of policy information. FIG. 5 is a schematic diagram illustrating an example of generating a feature vector. FIG. 6 is a schematic diagram illustrating an example of feature extraction. FIG. 7 is a diagram illustrating examples of similar or dissimilar existing policies. FIG. 8 is a diagram illustrating examples of similar or dissimilar existing policies. FIG. 9 is a diagram illustrating an example of an evaluation value graph. FIG. 10 is a diagram illustrating an example of a display on a client terminal. FIG. 11 is a flowchart illustrating the procedure for outputting similar policies. FIG. 12 is a schematic diagram illustrating an example of a machine learning model. FIG. 13 is a flowchart illustrating the procedure for outputting similar policies. FIG. 14 is a diagram illustrating an example of a hardware configuration.

[0011] Hereinafter, with reference to the accompanying drawings, a description will be given of modes for carrying out a policy output method, an information processing device, and a policy output program (hereinafter referred to as "embodiments") according to the present application. Each embodiment merely illustrates examples and aspects, and such examples do not limit the range of values, functions, or usage scenarios. Furthermore, each embodiment can be adaptively combined within a range that does not cause inconsistencies in the processing content.

[0012] <First Embodiment> <System Configuration> Fig. 1 is a block diagram showing an example of the functional configuration of a server device 10. The server device 10 shown in Fig. 1 provides a data-based platform that enables sharing, cross-referencing, and updating of policy flow data.

[0013] For example, the server device 10 can provide the functions of the data infrastructure platform as a cloud service by executing PaaS (Platform as a Service) type middleware or SaaS (Software as a Service) type applications.

[0014] As shown in Fig. 1, the 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 above-mentioned data infrastructure. For example, the client terminal 30 may be used by a policy planner as an example of an entity that implements a policy, such as a local government or other related party. Note that 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] <Flow Graph of Policy> An example of a flow graph of the above policy is shown in Figure 2. Figure 2 is a diagram illustrating an example of a flow graph of a policy. Z1, Z2, Z3, and Z4 in Figure 2 indicate, for example, services that an administrator provides to a user. These may also be referred to as "service implementation components." Specific examples of services include, in the medical field, "interventions" to which the target object of the policy, such as a resident, 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.

[0017] 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 conditions in the medical field.

[0018] 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."

[0019] In this embodiment, policy planning in the medical field will be described as an example, but the present invention is not limited to this. The above-described embodiment may be used for various policy planning such as work with conditional branching, tests, and questionnaires. In this case, the same effects as those of the above-described embodiment can be obtained.

[0020] Figure 3 shows 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 on the flow of people and parameters based on actual values ​​when each conditional branching component is used. This allows local governments to utilize policies implemented by other local governments that are suitable for their own purposes, taking into account their own resources.

[0021] 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 "no intervention" by a specialist is required for the citizen, as shown at S5.

[0022] On the other hand, if "eGFR<α" is satisfied (see the YES route at S2), then component #3 as conditional branch component C is set to "HbA1c<β" as shown at S3. If "HbA1c<β" is satisfied (see the YES route at S3), then component #4 as conditional branch component D is set to "nephrologist" as shown at 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 S3), then it is determined that the citizen requires intervention by a "diabetes specialist" as shown at S7.

[0023] In the example shown in Figure 3, the number of people who will flow through part #1, part #2, part #3, and part #4 in that order is predicted, as indicated by the arrows. For example, in the policy flow graph shown in Figure 3, the results of assigning the number of people N = 1000 to interventions Z2 to Z4 are as follows: 50 people are assigned to intervention Z2. 150 people are assigned to intervention Z3. Furthermore, 800 people are assigned to intervention Z4.

[0024] 2 and 3, a specific example of another use of the policy flow graph will be described. The server device 10 searches the policy flow graph using attribute information of people in the organization to which the policy planner belongs. The server device 10 identifies the node to which the person is classified among the nodes located at the end of the policy flow graph. This allows the local government that applied the policy to identify medical institutions to recommend to the person, taking into account the health status of the person belonging to the local government and the local government's resources.

[0025] First, the server device 10 identifies people in the organization to which the policy planner belongs. For example, the server device 10 identifies residents of a local government. Next, the server device 10 uses attribute information of people in the organization to which the policy planner belongs to search the output policy flow graph to identify a node to which the identified person is classified among the nodes located at the end of the policy flow graph. The attribute information is biometric information identified by analyzing the person's bodily fluids. The attribute information includes estimated glomerular filtration rate, hemoglobin A1c level, urinary protein level, etc. Note that bodily fluids include blood, lymph, tissue fluid (interstitial fluid, intercellular fluid, interstitial fluid), sweat, tears, nasal mucus, urine, semen, vaginal fluid, amniotic fluid, milk, etc.

[0026] At this time, the server device 10 identifies a node into which the person is classified by comparing the person's attribute information with the conditions included in the conditional branch component. The server device 10 identifies a node into which the identified person is classified among the terminal nodes. The server device 10 then sets the medical institution indicated by the identified node into which the person is classified as a medical institution to be recommended to the identified person. The medical institution indicated by the node is a nephrologist, a diabetes specialist, etc.

[0027] Hereinafter, the flow graph of a policy may be abbreviated as "policy flow." Furthermore, among policy flows, a policy flow corresponding to a draft may be referred to as an "original policy flow," while a policy flow corresponding to an existing policy may be referred to as an "existing policy flow." Note that the "original policy" here refers to a policy designated as a draft when the policy is formulated. For example, it may be possible to designate one of the existing policies as is, or it may be possible to designate a modified policy in which a portion of an existing policy has been modified, or it may be possible to designate a newly created policy.

[0028] <Data Infrastructure> In the above data infrastructure, policy flows may be shared in any framework. As just one example, the above data infrastructure allows policy flows to be shared among organizations around the world, for example, public organizations such as local governments.

[0029] A planner can refer to templates of existing plans from around the world collected in the data base via the client terminal 30. For example, the planner can update the original plan by incorporating all or part of an existing plan similar to the original plan from the templates collected in the data base.

[0030] Thus, when drafting 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. For this reason, it is becoming increasingly important to compare the flow graph of the proposed policy with the flow graph of existing policies that serve as reference.

[0031] <Configuration of Server Device 10> Fig. 1 shows a schematic diagram of blocks related to the data infrastructure 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 a selection of functional units related to the data infrastructure, and the server device 10 may also be provided with functional units other than those shown.

[0032] The communication control unit 11 is a functional unit that controls communication with other devices such as the client terminal 30. As just one example, 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 to register policy information including a policy flow from the client terminal 30 or a request to output a policy similar to the original plan, or outputs a policy similar to the original plan to the client terminal 30.

[0033] The storage unit 13 is a functional unit that stores various types of data. As an example, the storage unit 13 is realized by internal, external, or auxiliary storage of the server device 10. For example, the storage unit 13 stores a policy DB (DataBase) 13A. The policy DB will be described together with the situation where reference, generation, or registration is performed.

[0034] 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. Alternatively, the control unit 15 can be realized by hardwired logic. As shown in FIG. 1 , the control unit 15 includes a receiving unit 15A, a registering unit 15B, an extracting unit 15C, a first calculating unit 15D, a second calculating unit 15E, and an output unit 15F.

[0035] The reception unit 15A is a processing unit that receives various requests from the client terminal 30. In one aspect, the reception unit 15A can receive a registration request for policy information including a policy flow from the client terminal 30. In another aspect, the reception unit 15A can receive a request for outputting policies similar to the original plan.

[0036] The registration unit 15B is a processing unit that registers the policy information in the policy DB 13A of the storage unit 13. As an example, when a registration request for policy information is accepted by the acceptance unit 15A, the registration unit 15B registers the policy information in the policy DB 13A.

[0037] FIG. 4 is a diagram illustrating an example of policy information. As shown in FIG. 4, the policy information received in the registration request may include the policy flow of an existing policy, the regional characteristics of the organization to which the policy planner of the policy flow belongs, for example, the local government, the evaluation value of the policy flow, the organization's resources, and other information. For example, the regional characteristics may include regional demographics, regional population distribution, regional traffic patterns, and facilities deployed in the region. The evaluation value may also include the actual values ​​of overall efficiency indicators and individual effectiveness indicators. Furthermore, the resources may include the scale and distribution of the number of medical personnel, medical facilities, etc. In addition, resources required for implementing the policy flow may also be included. The other information may include a machine learning model that inputs the policy flow, regional characteristics, etc., and outputs an evaluation value. A collection of policy information for such existing policies is managed by the facility DB.

[0038] The extraction unit 15C is a processing unit that extracts feature quantities related to the policy flow and regional characteristics. As an example, the extraction unit 15C can start processing when the reception unit 15A receives an output request for a policy similar to the original policy. The original policy α specified in the output request for such a similar policy may be selected from existing policies registered in the policy DB 13A, or a policy not registered in the policy DB 13A may be selected from the client terminal 30.

[0039] For example, the extraction unit 15C extracts feature quantities of policy flow and regional characteristics for each draft specified in the output request for the similar policy and for each existing policy stored in the policy DB 13A. FIG. 5 is a schematic diagram showing an example of generating feature vectors. Examples of regional characteristics in FIG. 5 include demographics, population distribution, facilities, and traffic patterns. As shown in FIG. 5, the extraction unit 15C extracts feature vectors in which the feature quantities representing the quantified features of policy flow, the feature quantities representing the quantified features of demographics, the feature quantities representing the quantified features of population distribution, and the feature quantities representing the quantified features of facilities and traffic patterns are expressed in vector format.

[0040] More specifically, the extraction unit 15C can generate a feature vector by performing feature extraction as shown in FIG. 6 from the policy flow and regional characteristics. FIG. 6 is a schematic diagram illustrating an example of feature extraction. As shown in FIG. 6, the extraction unit 15C can extract, as feature quantities of the policy flow, the number of nodes, the depth of the flow, the number of branches, the number of interventions, the number of types of interventions, and the like. Furthermore, the extraction unit 15C can extract, as feature quantities of demographics, statistical values ​​such as the mean and variance of age distribution, and statistical values ​​such as the mean and variance of mortality distribution. Furthermore, the extraction unit 15C can extract, as feature quantities of facilities, the number of hospitals or wards by medical function, such as acute phase or convalescent phase. Furthermore, the extraction unit 15C can extract, as feature quantities of population distribution, statistical values ​​of population density distribution, such as the total population, elderly population, young population, and working-age population, such as the mean, variance, skewness, kurtosis, median, maximum value, and minimum value.

[0041] Returning to the explanation of FIG. 1 , the first calculation unit 15D is a processing unit that calculates the similarity between the original plan and the existing measures. As just one example, the first calculation unit 15D calculates the similarity, such as the inner product or cosine similarity, between the feature vector extracted from the original plan by the extraction unit 15C and the feature vector extracted for each of the N existing measures by the extraction unit 15C. Note that the similarity may be calculated by normalizing the distance calculated between the feature vectors of the original plan and the existing measures.

[0042] 7 and 8 are diagrams showing examples of similar and dissimilar existing policies. FIG. 7 shows existing policies f1, which have a graph structure similar to the original policy flow F1, and existing policies f2 and f3, which have graph structures dissimilar to the original policy flow F1. FIG. 8 shows existing policies with regional characteristics d1 similar to the original policy characteristic D1 and existing policies with regional characteristics d2 dissimilar to the original policy characteristic D1. The regional characteristics D1, d1, and d2 shown in FIG. 8 may include the age distribution, mortality distribution, working population distribution, population density distribution, number of hospitals, total length of arterial roads, number of arterial road branches, and the like, shown in FIG. 6.

[0043] As shown in Figure 7, existing policy f1, which has a similar graph structure to the original policy flow F1, has a closer distance between the feature quantities of the policy flows, which increases the similarity to the original policy flow F1, making it more likely to be a similar policy. On the other hand, existing policies f2 and f3, which have dissimilar graph structures to the original policy flow F1, have a greater distance between the feature quantities of the policy flows, which decreases the similarity to the original policy flow F1, making them more likely to be dissimilar policies.

[0044] 8, an existing policy with a regional characteristic d1 similar to the original regional characteristic D1 is likely to be a similar policy because the distance between the regional characteristic features is close, and therefore the similarity to the original policy is high. On the other hand, an existing policy with a regional characteristic d2 dissimilar to the original regional characteristic D1 is likely to be a dissimilar policy because the distance between the regional characteristic features is farther, and therefore the similarity to the original policy is low.

[0045] Returning to the description of FIG. 1 , the second calculation unit 15E is a processing unit that calculates intervals of evaluation values ​​of similar policies of the original plan. As just one example, the second calculation unit 15E extracts, from among the N existing policies included in the policy DB 13A, existing policies whose similarity calculated by the first calculation unit 15D is equal to or greater than a threshold, as similar policies. Next, the second calculation unit 15E references the evaluation values ​​of M (≦N) similar policies whose similarity is equal to or greater than the threshold. In this situation where the evaluation values ​​of the M similar policies have been referenced, the second calculation unit 15E can calculate maximum and minimum intervals of the evaluation values ​​and calculate confidence intervals corresponding to the similarity distribution of the evaluation values.

[0046] In one aspect, the second calculation unit 15E calculates the maximum and minimum evaluation values ​​corresponding to each of the K evaluation axes among the evaluation values ​​of the M similar policies. The maximum and minimum evaluation values ​​calculated for each of the K evaluation axes in this manner can define the maximum and minimum intervals of the evaluation values ​​in the K-dimensional evaluation space.

[0047] In another aspect, the second calculation unit 15E calculates confidence intervals, such as 90% confidence intervals or 95% confidence intervals, for the M similar measures based on the distribution of the similarities of the M similar measures for each of the K evaluation axes, thereby making it possible to set confidence intervals in the K-dimensional evaluation space that exclude outliers and abnormal values ​​of the evaluation values.

[0048] The output unit 15F is a processing unit that outputs various information to the client terminal 30. In one aspect, the output unit 15F can output to the client terminal 30 the evaluation values ​​of M similar policies, among the N existing policies included in the policy DB 13A, whose similarities calculated by the first calculation unit 15D are equal to or greater than a threshold. In this case, the output unit 15F can generate an evaluation value graph by plotting the evaluation values ​​of the M similar policies corresponding to each evaluation axis. Furthermore, the output unit 15F can plot the maximum and minimum intervals of the evaluation values ​​calculated by the second calculation unit 15E on the evaluation value graph, or plot the confidence intervals of the M similar policies calculated by the second calculation unit 15E.

[0049] FIG. 9 is a diagram illustrating an example of an evaluation value graph. In FIG. 9, an index of overall efficiency (bed occupancy rate) and an index of individual effectiveness (maximum length of hospital stay) are illustrated as examples of evaluation axes. As shown in FIG. 9, evaluation value graph G1 and evaluation value graph G2 plot the evaluation values ​​(actual values) of M similar measures corresponding to each of the two evaluation axes. Of these, evaluation value graph G1 plots the maximum and minimum intervals of the evaluation values, i.e., dashed-line frames. Meanwhile, evaluation value graph G2 plots the similarities of the M similar measures associated with the evaluation values ​​of the M similar measures for each of the two evaluation axes, and also plots the confidence intervals of the M similar measures, i.e., elliptical regions. These evaluation value graphs G1 and G2 enable evaluation of untested measures based on actual values, thereby improving the evaluation accuracy of untested measures.

[0050] In another aspect, the output unit 15F can output a list of M similar policies whose similarities calculated by the first calculation unit 15D are equal to or greater than a threshold. Furthermore, the output unit 15F can narrow down the M similar policies to those whose evaluation values ​​satisfy a predetermined condition, such as those with the best value or a specific number of similar policies, and output the similar policies whose evaluation values ​​satisfy the predetermined condition as proposed policies. Additionally, the output unit 15F can output the M similar policies, or similar policies whose evaluation values ​​satisfy the predetermined condition, whose required resources for implementation satisfy a predetermined condition, such as those within an allowable range set based on the resources available in the local government of the policy planner.

[0051] 10 is a diagram showing an example display of the client terminal 30. As shown in FIG. 10, the window 200 displays the evaluation value graph G1 shown in FIG. 9. Furthermore, the window 200 displays the policy flow of the original plan, as well as the policy flows of M similar policies whose similarities are equal to or greater than a threshold, and a list of the evaluation values ​​of the M similar policies by evaluation axis. Furthermore, the window 200 displays the similar policy with the best evaluation value among the M similar policies as the proposed policy. Using this window 200, existing policies that have regional characteristics similar to those of the original plan can be output as similar policies.

[0052] 10 shows an example in which a list of M similar measures is displayed, but it is also possible to display the action flow of similar measures designated by the user from the evaluation value graph G1 or the evaluation value graph G2. In addition, it is also possible to display the similar measures with the smallest evaluation value on each evaluation axis within the maximum and minimum evaluation value intervals of the evaluation value graph G1, or the similar measures with the largest evaluation value on each evaluation axis. Furthermore, it is also possible to display the action flow of the similar measures located closest to the center or center of gravity of the similarity distribution included in the evaluation value graph G2.

[0053] 11 is a flowchart showing the procedure of the output process of similar measures. This process can be started when the receiving unit 15A receives a request to output similar measures of the original plan.

[0054] 11 , when the receiving unit 15A receives an output request for measures similar to the original plan α (step S101), the extraction unit 15C executes the following process. That is, the extraction unit 15C extracts feature vectors in which the feature amounts of the measure flow of the original plan α are quantified and the feature amounts of the regional characteristics of the original plan α are expressed in vector format (step S102). Next, the extraction unit 15C acquires N existing measures included in the measure DB 13A stored in the storage unit 13 (step S103).

[0055] Thereafter, loop processing 1 is executed in which the processing of step S104 and step S105 described below are repeated a number of times corresponding to the number N of existing policies acquired in step S103.

[0056] That is, the extraction unit 15C extracts a feature vector in which the feature quantities that represent the characteristics of the policy flow of the nth existing policy and the feature quantities that represent the characteristics of the regional characteristics of the nth existing policy are expressed in vector format (step S104).

[0057] Then, the first calculation unit 15D calculates the similarity between the feature vector of the original plan α extracted in step S102 and the feature vector of the n-th existing policy extracted in step S104 (step S105).

[0058] By repeating this loop process 1, the similarity is calculated for each of the N existing policies.

[0059] After that, the second calculation unit 15E sorts the N existing policies in descending order of the similarity calculated in step S105 (step S106), and then extracts the top M existing policies with the highest similarity (step S107).

[0060] The second calculation unit 15E then calculates intervals of the evaluation values ​​(actual values) of the M existing policies, such as maximum / minimum intervals or confidence intervals (Step S108). Subsequently, the output unit 15F further plots the intervals of the evaluation values ​​calculated in Step S108 on an evaluation value graph on which the evaluation values ​​of the M similar policies corresponding to each evaluation axis are plotted, and displays the plot on the client terminal 30 (Step S109).

[0061] Furthermore, the output unit 15F displays on the client terminal 30 as proposed measures those similar measures among the M similar measures whose evaluation values ​​satisfy predetermined conditions, for example, similar measures that correspond to the best value or a top specified number (step S110), and terminates the processing.

[0062] <One Aspect of Effect> As described above, when a flow graph of an original plan is specified, the server device 10 according to this embodiment calculates the similarity of the flow graph between the original plan and existing measures, and outputs the evaluation value stored in association with a similar measure among the flow graphs of the existing measures whose similarity is equal to or greater than a threshold. This allows inexperienced measures to be evaluated based on performance values. Therefore, the server device 10 according to this embodiment can improve the accuracy of evaluating inexperienced measures.

[0063] Furthermore, when a flow graph of an original plan is specified, the server device 10 according to this embodiment calculates the similarity of regional characteristics between the original plan and existing measures, and outputs the flow graph of a similar measure among the flow graphs of the existing measures whose similarity is equal to or greater than a threshold value. Therefore, according to the server device 10 according to this embodiment, it is possible to output, as a similar measure, an existing measure that has regional characteristics similar to those of the original plan.

[0064] Although the embodiments relating to the disclosed device have been described above, the present invention may be embodied in various different forms other than the above-described embodiments. Therefore, other embodiments included in the present invention will be described below.

[0065] <Machine Learning Model for Predicting Evaluation Values> The number of draft plans created by a policy planner is not necessarily limited to one. For example, a scenario in which K draft plans are created by partially rearranging one or more existing policies can be considered. In this case, as the number of draft plans increases, the processing load shown in FIG. 11 also increases. From this perspective, a machine learning model that outputs evaluation values ​​using policy flow, regional characteristics, and the like as input can be used to narrow down the number of draft plans to be compared with existing policies.

[0066] 12 is a schematic diagram showing an example of a machine learning model. As shown in FIG. 12, a machine learning model m is used to predict the evaluation values ​​of the policy flows of K original plans. For example, the machine learning model m may be realized using a neural network, a support vector machine, gradient boosting, or the like. To train such a machine learning model m, a dataset TR can be used, which includes training data in which the feature values ​​of at least one of the policy flows, regional characteristics, and resources are associated with the correct labels of the evaluation values.

[0067] For example, in the training phase, at least one feature quantity from among the policy flow, regional characteristics, and resources can be used as an explanatory variable of the machine learning model m, and the label can be used as the objective variable of the machine learning model m, and the machine learning model m can be trained according to any machine learning algorithm, such as deep learning, thereby obtaining a trained machine learning model M.

[0068] In the prediction phase, at least one feature of the original plan's policy flow, regional characteristics, and resources is input to the machine learning model M. The machine learning model M to which the feature has been input outputs an evaluation value of the original plan's policy flow.

[0069] Fig. 13 is a flowchart showing the procedure for outputting similar measures. In the flowchart shown in Fig. 13, the same step numbers are assigned to steps in which the same processes as those in the flowchart shown in Fig. 11 are executed.

[0070] 13, when the receiving unit 15A receives a request to output K similar measures for the original plans (step S201), the following process is executed: That is, a loop process 0 is executed in which the process of step S202 is repeated a number of times corresponding to the number K of the original plans.

[0071] For example, at least one feature of the policy flow, regional characteristics, and resources of the kth draft is input to the machine learning model M (step S202), which then outputs a predicted evaluation value of the kth draft.

[0072] By repeating this loop process 0, a predicted evaluation value is obtained for each of the K original plans. Then, the extraction unit 15C extracts original plan α from the K original plans, whose predicted evaluation value satisfies a predetermined condition, for example, the best value (step S203).

[0073] The extraction unit 15C then extracts feature vectors in which the feature quantities that represent the quantified characteristics of the policy flow of the original plan α extracted in step S203 and the feature quantities that represent the quantified characteristics of the regional characteristics of the original plan α are expressed in vector form (step S102). Subsequently, the extraction unit 15C acquires N existing policies included in the policy DB 13A stored in the storage unit 13 (step S103).

[0074] Thereafter, loop processing 1 is executed in which the processing of step S104 and step S105 described below are repeated a number of times corresponding to the number N of existing policies acquired in step S103.

[0075] That is, the extraction unit 15C extracts a feature vector in which the feature quantities that represent the characteristics of the policy flow of the nth existing policy and the feature quantities that represent the characteristics of the regional characteristics of the nth existing policy are expressed in vector format (step S104).

[0076] Then, the first calculation unit 15D calculates the similarity between the feature vector of the original plan α extracted in step S102 and the feature vector of the n-th existing policy extracted in step S104 (step S105).

[0077] By repeating this loop process 1, the similarity is calculated for each of the N existing policies.

[0078] After that, the second calculation unit 15E sorts the N existing policies in descending order of the similarity calculated in step S105 (step S106), and then extracts the top M existing policies with the highest similarity (step S107).

[0079] The second calculation unit 15E then calculates intervals of the evaluation values ​​(actual values) of the M existing policies, such as maximum / minimum intervals or confidence intervals (Step S108). Subsequently, the output unit 15F further plots the intervals of the evaluation values ​​calculated in Step S108 on an evaluation value graph on which the evaluation values ​​of the M similar policies corresponding to each evaluation axis are plotted, and displays the plot on the client terminal 30 (Step S109).

[0080] Furthermore, the output unit 15F displays on the client terminal 30 as proposed measures those similar measures among the M similar measures whose evaluation values ​​satisfy predetermined conditions, for example, similar measures that correspond to the best value or a top specified number (step S110), and terminates the processing.

[0081] As described above, by executing the processes from step S201 to step S203, the number of original plans to be compared with existing measures can be narrowed down using a machine learning model for predicting evaluation values.

[0082] Here, we have given an example of using a machine learning model for predicting evaluation values ​​to narrow down the number of original plans to be compared with existing measures, but the predicted evaluation value can also be used to determine the threshold value of the evaluation value of similar measures from which the second calculation unit 15E will extract confidence intervals.

[0083] <Application Example of Feature Quantities> In the first embodiment described above, an example of generating a feature vector including feature quantities of a policy flow and regional characteristics has been given, but the present invention is not limited to this. For example, the feature vector may include, as feature quantities of evaluation values, statistical values ​​such as the mean and variance of BI score differences in a region, statistical values ​​such as the mean and variance of length of hospital stay in a region, and statistical values ​​such as the mean and variance of bed occupancy rates in a region. Furthermore, the feature vector may include, as feature quantities of resources, statistical values ​​such as the mean and variance of the number of medical personnel in a region, and statistical values ​​such as the mean and variance of the number of medical resources in a region (ambulances, helicopters, intensive care units (ICUs), magnetic resonance imaging (MRIs), etc.).

[0084] <Distribution and Integration> Furthermore, the components of each device shown in the figure 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 the figure, and all or part of the devices can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. For example, the reception unit 15A, registration unit 15B, extraction unit 15C, first calculation unit 15D, second calculation unit 15E, and output unit 15F may be connected via a network as external devices to the server device 10. Alternatively, the reception unit 15A, registration unit 15B, extraction unit 15C, first calculation unit 15D, second calculation unit 15E, and output unit 15F may be included in separate devices, each of which may be connected via a network to cooperate with the other devices to realize the functions of the server device 10.

[0085] <Hardware Configuration> The various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer, a workstation, etc. Therefore, an example of a computer that executes a policy output program having the same functions as those in the first and second embodiments will be described below with reference to FIG.

[0086] Fig. 14 is a diagram showing an example of a hardware configuration. As shown in Fig. 14, a computer 100 has an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. The computer 100 also has a CPU 150, a ROM 160, a HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

[0087] As shown in Fig. 14, the HDD 170 stores a policy output program 170a that performs the same functions as the reception unit 15A, registration unit 15B, extraction unit 15C, first calculation unit 15D, second calculation unit 15E, and output unit 15F shown in the first embodiment. This policy output program 170a may be integrated or separated, similar to the respective components of the reception unit 15A, registration unit 15B, extraction unit 15C, first calculation unit 15D, second calculation unit 15E, and output unit 15F shown in Fig. 1. In other words, the HDD 170 does not necessarily have to store all of the data shown in the first embodiment, as long as the data used for processing is stored in the HDD 170.

[0088] Under such an environment, the CPU 150 reads the policy output program 170a from the HDD 170 and loads it into the RAM 180. As a result, the policy output program 170a functions as a policy output process 180a, as shown in FIG. 14. The policy output process 180a loads various data read from the HDD 170 into an area of ​​the storage area of ​​the RAM 180 allocated to the policy output process 180a, and executes various processes using the loaded data. For example, examples of processes executed by the policy output process 180a include the processes shown in FIGS. 11 and 13. Note that the CPU 150 does not necessarily need to operate all of the processing units shown in the first embodiment above; it is sufficient that the processing units corresponding to the processes to be executed are virtually implemented.

[0089] The policy output program 170a does not necessarily have to be stored in the HDD 170 or the ROM 160 from the beginning. For example, each program may be stored on a "portable physical medium" such as a flexible disk, a so-called FD, a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card that is inserted into the computer 100. The computer 100 may then retrieve and execute each program from such a portable physical medium. Alternatively, each program may be stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, or the like, and the computer 100 may retrieve and execute each program from such a computer or server device.

[0090] REFERENCE SIGNS LIST 10 Server device 11 Communication control unit 13 Storage unit 13A Policy DB 15 Control unit 15A Reception unit 15B Registration unit 15C Extraction unit 15D First calculation unit 15E Second calculation unit 15F Output unit 30 Client terminal

Claims

1. The computer Obtain the regional characteristics of the first region to which the policy will be applied, referencing a storage unit that stores other measures applied to each of a plurality of second regions and regional characteristics of each of the plurality of second regions, and calculating a similarity between the regional characteristics of the first region and each of the regional characteristics of the plurality of second regions; outputting a measure applied to a similar area whose similarity is equal to or greater than a threshold value from among the plurality of other measures stored in the storage unit; A policy output method characterized by executing a process.

2. The policy output method described in claim 1, characterized in that the process of calculating the similarity includes, when obtaining regional characteristics of a first region to which a policy flow including conditional branches and nodes connected by directed edges is to be applied, referring to a memory unit that stores other policies applied to each of multiple second regions and the regional characteristics of each of the multiple second regions, and calculating the similarity between the regional characteristics of the first region and each of the regional characteristics of the multiple second regions.

3. The policy output method described in claim 2, characterized in that the output process includes a process of referring to the memory unit that further stores evaluation values ​​indicating the number of targets assigned to a node by each of the other policies in each of the plurality of second regions, and outputting a policy flow including the conditional branch and the node that is a policy in a similar region among the other policies in each of the plurality of second regions and whose evaluation value satisfies a predetermined condition.

4. The policy output method described in claim 3, characterized in that the output process includes a process of referring to the memory unit that further stores the resources necessary for implementing the other policies in each of the plurality of second regions, and outputting a policy flow of policies in similar regions among the other policies in each of the plurality of second regions, whose resources satisfy specified conditions.

5. Identify the people in the organization involved in the measure; By searching the output policy flow using attribute information of the person included in the organization, a node to which the identified person is classified is identified among the nodes located at the end of the policy flow; setting the medical institution indicated by the identified classified node as a medical institution to be recommended to the identified person; 5. The method for outputting measures according to claim 4.

6. the attribute information of the person is biometric information identified by analyzing the body fluid of the person; the conditional branch includes a condition regarding at least one of an estimated glomerular filtration rate, a hemoglobin A1c value, and a urinary protein value; the terminal nodes include nodes indicating at least one of a nephrologist and a diabetologist, The policy output method described in claim 5, characterized in that the node into which the identified person is classified is identified among the nodes located at the end by comparing attribute information of a person belonging to an organization related to the planner of the policy with the conditional branch.

7. 2. The method for outputting measures according to claim 1, wherein the regional characteristics are at least one of population distribution, demographics, traffic patterns, and facility deployment in the region.

8. Obtain the regional characteristics of the first region to which the policy will be applied, referencing a storage unit that stores other measures applied to each of a plurality of second regions and regional characteristics of each of the plurality of second regions, and calculating a similarity between the regional characteristics of the first region and each of the regional characteristics of the plurality of second regions; outputting a measure applied to a similar area whose similarity is equal to or greater than a threshold value from among the plurality of other measures stored in the storage unit; An information processing device comprising a control unit that executes processing.

9. On the computer, Obtain the regional characteristics of the first region to which the policy will be applied, referencing a storage unit that stores other measures applied to each of a plurality of second regions and regional characteristics of each of the plurality of second regions, and calculating a similarity between the regional characteristics of the first region and each of the regional characteristics of the plurality of second regions; outputting a measure applied to a similar area whose similarity is equal to or greater than a threshold value from among the plurality of other measures stored in the storage unit; A policy output program characterized by executing a process.