Decision support device, decision support method, and decision support program

Abstract

This decision support device executes: acquisition processing for acquiring a feature amount related to a measure to be taken by a group, a feature amount related to an action of the group, and a feature amount related to the group; first prediction processing for inputting the feature amount related to the measure, the feature amount related to the action, and the feature amount related to the group to a machine learning model, the feature amounts being acquired by the acquisition processing, and outputting a first prediction result of prediction of the action in cooperation among members constituting the group, the machine learning model predicting whether to perform the action in cooperation among members constituting the group when the feature amount related to the action and the feature amount related to the group are input depending upon whether the measure has been implemented; and outputting processing for outputting the first prediction result obtained by the first prediction processing.

Description

Decision-making support device, decision-making support method, and decision-making support program

[0001] The present invention relates to a decision-making support device, a decision-making support method, and a decision-making support program for processing information.

[0002] Patent Document 1 below discloses an organization improvement activity support device that improves the continuity of organization improvement activities without on-site expert intervention. This organization improvement activity support device includes: a policy conformance acquisition means for acquiring a policy conformance that quantitatively indicates the degree of conformance of a policy being implemented by a target organization with the characteristics and improvement theme of the target organization; a policy agreement acquisition means for acquiring a policy agreement that quantitatively indicates the degree of agreement of members of the target organization with the policy; a policy activity acquisition means for acquiring a policy activity that quantitatively indicates the degree of implementation of the policy in the target organization; an effect point calculation means for calculating effect points that quantitatively indicate the degree of effectiveness of the organization improvement activities in the target organization based on the policy conformance, the policy agreement, and the policy activity; and an information presentation means for presenting information on at least the activity status of the target organization to facilitators and / or members of the target organization based on the effect points.

[0003] Japanese Patent Application Laid-Open No. 2016-139323

[0004] The measures proposed based on the characteristics of the organization are not concrete measures, and the consideration of concrete measures requires manpower.

[0005] The present invention aims to automatically present the cooperation rate among members of a group when a policy for promoting behavior is implemented in the group.

[0006] A decision-making support device according to a first aspect of the invention disclosed in the present application is a decision-making support device having a processor that executes a program and a storage device that stores the program, wherein the processor executes an acquisition process to acquire features related to a measure to be taken by a group, features related to the behavior of the group, and features related to the group; a first prediction process in which, when the features related to the behavior and the features related to the group are input depending on whether the measure is implemented or not, the features related to the measure, the features related to the behavior, and the features related to the group acquired by the acquisition process are input into a machine learning model that predicts whether members of the group will cooperate to take the action, and outputs a first prediction result that predicts the behavior of the members of the group cooperate; and an output process that outputs the first prediction result obtained by the first prediction process.

[0007] A decision support apparatus according to a second aspect of the invention disclosed in this application comprises a processor that executes a program and a storage device that stores the program, and is capable of communicating with a plurality of computers, wherein the decision support apparatus holds master information in which recommended actions for a group are ranked according to the characteristics of the group, and each of the plurality of computers holds slave information in which recommended actions for the group are ranked according to the characteristics of the group, and the processor calculates a ranking based on the specific characteristics of the group, the specific recommended actions for the group, and a first answer to a question given to each member of the group before the implementation of a measure to be taken by the group. and a second score for the group based on a second answer to the question given to each of the members after the measure has been implemented, from each of the plurality of computers; a calculation process for calculating a correction value for a ranking of a specific recommended action for the group for a specific characteristic of the group based on the first score and the second score received by the receiving process; a correction process for correcting the ranking of the specific recommended action based on the correction value calculated by the calculation process; and a transmission process for transmitting the correction result by the correction process to the plurality of computers.

[0008] A decision support device according to a third aspect of the invention disclosed in the present application is a decision support device that has a processor that executes a program and a storage device that stores the program, and is capable of communicating with a plurality of computers, wherein the decision support device holds master information in which a ranking of recommended actions for a group is set according to the characteristics of the group, and each of the plurality of computers holds a machine learning master model that, when a feature related to the action of the group and a feature related to the group are input, predicts whether members of the group will cooperate to take the action, depending on whether a measure to be taken by the group is implemented, and each of the plurality of computers predicts whether the member of the group will cooperate to take the action, depending on whether the measure is implemented. The system maintains a machine learning slave model that, when inputted with features related to behavior and features related to the group, predicts whether members of the group will cooperate to perform the behavior, and the processor executes a receiving process that receives information related to the learning of the machine learning master model in the machine learning slave model from each of the plurality of computers, a learning process that learns the machine learning master model based on the information related to the learning received from each of the plurality of computers by the receiving process, and a transmitting process that transmits learning parameters of the machine learning master model learned by the learning process to the plurality of computers.

[0009] According to a representative embodiment of the present invention, when a policy for promoting behavior is implemented in a group, the cooperation rate among members of the group can be automatically displayed. Problems, configurations, and effects other than those described above will become clear from the description of the following examples.

[0010] FIG. 1 is a block diagram showing an example of the hardware configuration of a decision-making support apparatus. FIG. 2 is an explanatory diagram showing an example of a member DB. FIG. 3 is an explanatory diagram showing an example of a response DB. FIG. 4 is an explanatory diagram showing an example of a group characteristic table. FIG. 5 is an explanatory diagram showing an example of a recommended action table. FIG. 6 is a flowchart showing an example of an information processing procedure performed by the decision-making support apparatus. FIG. 7 is an explanatory diagram showing an example of a group analysis result obtained by the group characteristic analysis process (step S602). FIG. 8 is an explanatory diagram showing an example of a recommended action analysis result. FIG. 9 is an explanatory diagram showing an example of a group data input screen. FIG. 10 is an explanatory diagram showing an example of the promotion measure analysis process (step S605). FIG. 11 is an explanatory diagram showing an example of a first data conversion table used in the data conversion (step S1001). FIG. 12 is an explanatory diagram showing an example of a second data conversion table used in the data conversion (step S1001). FIG. 13 is an explanatory diagram showing an example of a reference feature 1011 output by the data conversion (step S1001). FIG. 14 is an explanatory diagram showing an example of a measure feature table. FIG. 15 is an explanatory diagram showing an example of addition (step S1002). FIG. 16 is an explanatory diagram showing an example of cooperation rate prediction (step S1003). FIG. 17 is an explanatory diagram showing an example of an input screen for promotion measure implementation results. FIG. 18 is an explanatory diagram showing an example of a decision making support system. FIG. 19 is an explanatory diagram showing an example of the survey implementation result group shown in FIG. 18. FIG. 20 is an explanatory diagram showing an example of update data. FIG. 21 is an explanatory diagram showing an example of a recommended action master table. FIG. 22 is an update sequence diagram for the recommended action table. FIG. 23 is an explanatory diagram showing an example of the promotion measure implementation result group shown in FIG. 18. FIG. 24 is an update sequence diagram for the machine learning model.

[0011] <Example of Hardware Configuration of Decision-Making Support Apparatus> Fig. 1 is a block diagram showing an example of the hardware configuration of a decision-making support apparatus. The decision-making support apparatus 100 includes a processor 101, a storage device 102, an input device 103, an output device 104, and a communication interface (communication IF) 105. The processor 101, the storage device 102, the input device 103, the output device 104, and the communication IF 105 are connected via a bus 106. The processor 101 controls the decision-making support apparatus 100. The storage device 102 serves as a working area for the processor 101. The storage device 102 is a non-transitory or temporary recording medium that stores various programs and data. Examples of the storage device 102 include a read-only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), and a flash memory. The input device 103 inputs data. Examples of the input device 103 include a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, a microphone, and a sensor. The output device 104 outputs data. Examples of the output device 104 include a display, a printer, and a speaker. The communication IF 105 connects to a network and transmits and receives data.

[0012] <Member DB> Figure 2 is an explanatory diagram showing an example of a member DB. The member DB 200 is a database that stores personal information of members. A member is a person (constituent) who belongs to an organization such as a company, government agency, or school, or who receives services from the organization. When multiple members gather, some kind of decision-making is carried out by the group made up of the multiple members.

[0013] The member DB 200 includes the fields member ID 201, age 202, gender 203, and nationality 204. Personal information other than age 202, gender 203, and nationality 204 may also be included. A combination of fields in the same row forms an entry that identifies the personal information of one member. Note that the value z of xxIDyyy (xx is a string, yyy is a code) may be expressed as xx#z. For example, a member with a member ID 201 value of "001" would be expressed as member #001.

[0014] <Response DB> Fig. 3 is an explanatory diagram showing an example of the response DB. The response DB 300 is a database that stores, for each group, responses to questions included in a questionnaire administered to each member belonging to the group. Fig. 3 shows example responses to questions from 35 members belonging to one group.

[0015] The response DB 300 has the fields of member ID 201 and A-type questions 301A to F-type questions 301F. When the A-type questions 301A to F-type questions 301F are not distinguished, they are referred to as X-type questions 301X. The A-type questions 301A are questions about autonomy. The B-type questions 301B are questions about adaptability. The C-type questions 301C are questions about creativity. The D-type questions 301D are questions about equality. The E-type questions 301E are questions about cooperation. The F-type questions 301F are questions about belonging.

[0016] Each of the A-series questions 301A to F-series questions 301F includes four questions (1) to (4). The questions (1) to (4) of the X-series question 301X are similar questions, and are intended to increase the reliability of the answers. Each of the questions (1) to (4) is scored by selecting one of the options "1" to "5."

[0017] Options "1" to "5" are scores on a five-point scale, where, for example, option "1" means "not at all applicable," option "2" means "not very applicable," option "3" means "applies," option "4" means "very applicable," and option "5" means "extremely applicable."

[0018] In FIG. 3, six A-series questions 301A to F-series questions 301F are shown as an example, but the number of questions is not limited to six. Also, X-series question 301X includes four questions (1) to (4), but the number of questions is not limited to four. Also, the options are not limited to a five-point scale. Also, the options may be reaction times to answers.

[0019] The answer DB 300 also includes a question score 302 for the X-type question 301X. The question score 302 is the average of the 35×4 scores for questions (1) to (4) of all 35 members belonging to the group. For example, the A-type question 301A indicates that the average of the 35×4 scores for questions (1) to (4) of the 35 members is "4.40."

[0020] The response DB 300 also includes a total score 303. The total score 303 is the average value of the 35 x 4 x 6 scores for questions (1) to (4) in the A-type questions 301A to F-type questions 301F of all 35 members belonging to the group. For example, in the case of the A-type questions 301A to F-type questions 301F, the average value of the 35 x 4 x 6 scores for questions (1) to (4) of the 35 members is "3.78."

[0021] Furthermore, although average values ​​are used for the question score 302 and the total score 303, statistical values ​​such as medians or average values ​​excluding outliers may also be used.

[0022] <Group Characteristics Table> Fig. 4 is an explanatory diagram showing an example of a group characteristics table 400. The group characteristics table 400 has the following fields: characteristic ID 401, comparison result with reference value 402, group type 403, and features and issues 404. A combination of values ​​in each field on the same row forms an entry that indicates one group characteristic.

[0023] The characteristic ID 401 is identification information that uniquely identifies a group characteristic. The comparison result 402 with the reference value indicates the result of comparing each question score 302 of the A-type questions 301A to the F-type questions 301F with the reference value. In the example of the 5-point scale, the score range is 1 to 5, so the question score 302 that serves as the reference value is set to "3.00", and if it is higher than the reference value, it is marked with "+", and if it is lower, it is marked with "-". The combination of "+" and "-" for the A-type questions 301A to the F-type questions 301F in the comparison result 402 with the reference value uniquely identifies the group type 403.

[0024] The group type 403 is a character string indicating the type of group characteristics. The characteristics and issues 404 are character strings indicating the characteristics and issues of the group corresponding to the group type 403.

[0025] 3, the question score 302 of the A-system question 301A is "4.40", the question score 302 of the B-system question 301B is "3.23", the question score 302 of the C-system question 301C is "4.05", ..., the question score 302 of the F-system question 301F is "1.55". If the question score 302 of the D-system question 301D is above the reference value of "3.00" and the question score 302 of the E-system question 301E is below the reference value of "3.00", then the group shown in FIG. 3 corresponds to the group characteristic of "G61".

[0026] 4, there are 64 types of group characteristics, but this is not limited to 64 types. Also, while the comparison result 402 with the reference value is expressed using two values, "+" and "-," the comparison result 402 with the reference value may be expressed using multiple values ​​based on the difference from the reference value. Also, the question scores 302 of the A-system questions 301A to F-system questions 301F may be used as vectors, and the standard value scores of the A-system questions 301A to F-system questions 301F may be used as standard value vectors, and the comparison result 402 with the standard value may be expressed as the distance between the two vectors.

[0027] Furthermore, the character strings shown in the group type 403 and the features and issues 404 are not limited to these, and other expressions may be used.

[0028] <Recommended Action Table> Fig. 5 is an explanatory diagram showing an example of a recommended action table. The recommended action table 500 is a table that specifies actions recommended for a group that corresponds to a group characteristic. The recommended action table 500 has a recommended action ID 501, a recommended action 502, and a group characteristic ranking 503.

[0029] The recommended action ID 501 is identification information that uniquely identifies the recommended action 502. The recommended action 502 includes actions recommended for the leader of the group and actions recommended for the members of the group. The recommended action ID 501 that uniquely identifies the recommended action 502 recommended for the leader has the suffix "L" added to the end, and the recommended action ID 501 that uniquely identifies the recommended action 502 recommended for the member has the suffix "M" added to the end.

[0030] The recommended action 502 is a character string indicating an action recommended for a group. The recommended action 502 includes a recommended action 502 recommended for the leader and a recommended action 502 recommended for the members. For example, the recommended action #R01L is "share the meaning of the group," and the recommended action #R01M is "participate in regular meetings."

[0031] The group characteristic ranking 503 indicates the ranking of the recommended actions 502 for each group characteristic. In this example, 20 recommended actions are defined, and therefore each recommended action 502 is ranked from 1st to 20th for each group characteristic.

[0032] The member DB 200, the response DB 300, the group characteristic table 400, and the recommended action table 500 are specifically stored in, for example, the storage device 102, but may also be stored in another computer accessible via a network such as the Internet, a LAN (Local Area Network), or a WAN (Wide Area Network) via the communication IF 105.

[0033] 6 is a flowchart showing an example of an information processing procedure by the decision-making support device 100. The decision-making support device 100 executes a response data acquisition process (step S601) and a group characteristic analysis process (step S602).

[0034] In the response data acquisition process (step S601), the decision-making support device 100 reads the question score 302 and the total score 303 as response data from the response DB 300. The decision-making support device 100 may read the scores of questions (1) to (4) of the X-series question 301X for each member from the response DB 300, and calculate the question score 302 and the total score 303 as response data.

[0035] [Group Characteristic Analysis Process (Step S602)] In the group characteristic analysis process (step S602), the decision-making support device 100 analyzes group characteristics based on the response data acquired by the response data acquisition process (step S601). Specifically, for example, the decision-making support device 100 refers to the group characteristic table 400, compares the question score 302 with a reference value, and identifies a comparison result 402 with the reference value. As a result, the characteristic ID 401, group type 403, and features and issues 404 of the group (hereinafter referred to as the target group) made up of members who provided the response data (question score 302) are identified as analysis information.

[0036] 7 is an explanatory diagram showing an example of a group analysis result obtained by the group characteristic analysis process (step S602). The group analysis result 700 can be displayed, for example, on a display, which is an example of the output device 104. The group analysis result 700 includes a radar chart 701, analysis information 702, an analysis button 703, and an end button 704.

[0037] The radar chart 701 visualizes the question scores 302 for each of the A-type questions 301A to F-type questions 301F. The first chart 711 is composed of six black dots indicating the question scores 302 for each of the A-type questions 301A to F-type questions 301F, and lines connecting adjacent black dots. The second chart 722 shows a reference value (for example, a question score of 3.00). The second chart 722 may also be a chart for another group.

[0038] The analysis information 702 is information that visualizes the characteristic ID 401, population type 403, and features and issues 404 of the target population.

[0039] The analysis button 703 and the end button 704 are user interfaces that can be pressed by a user (for example, a leader of a target group) by operating the input device 103. When the analysis button 703 is pressed (step S603: Yes), a recommended behavior analysis process (step S604) is executed. When the end button 704 is pressed (step S603: No), the series of processes ends.

[0040] The group characteristic analysis process (step S602) can also be realized by machine learning using the response data as feature quantities and the group type 403 as correct answer data, rather than the rule-based method described above.

[0041] In FIG. 6, the decision support apparatus 100 executes a recommended behavior analysis process (step S604), a promotion measure analysis process (step S605), and a measure implementation process (step S606).

[0042] [Recommended Action Analysis Process (Step S604)] In the recommended action analysis process (step S604), the decision-making support device 100 analyzes the recommended actions 502. Specifically, for example, the decision-making support device 100 acquires the characteristic ID 401 and the character string of the issue in the features and issues 404 from the analysis information 702. Furthermore, the decision-making support device 100 acquires, from the recommended action table 500, the top N recommended actions 502 in the group characteristic ranking 503, among the recommended actions 502 for the group characteristics identified in the group characteristic analysis process (step S602).

[0043] 8 is an explanatory diagram showing an example of a recommended behavior analysis result. The recommended behavior analysis result 800 has a task display area 801, a top recommended behavior display area 802, and a selection button 803. The recommended behavior analysis result 800 can be displayed, for example, on a display, which is an example of the output device 104.

[0044] The problem display area 801 is an area that displays the characteristic ID 401 of the target group, its features, and the character string of the problem among the problems 404. In this example, since the acquired characteristic ID 401 is "G61", "Low sense of belonging, so prone to drop out" is displayed.

[0045] The top recommended action display area 802 is an area that displays the top N recommended actions 502 in the group characteristic ranking 503 obtained from the recommended action table 500. In this example, N=4. When the characteristic ID 401 is "G61", the recommended actions 502 displayed are #R02 (1st place), #R07 (2nd place), #R20 (3rd place), and #R13 (4th place).

[0046] The recommended actions 821 for the leader side are recommended actions #R02L, #R07L, #R20L, and #R13L, and the recommended actions 822 for the member side are recommended actions #R02M, #R07M, #R20M, and #R13M. The recommended actions 822 for the member side include check boxes 823. The check boxes 823 are a user interface that can be selected by a user (for example, the leader of the target group) by operating the input device 103.

[0047] The selection button 803 is a user interface that can be pressed by a user (for example, the leader of the target group) by operating the input device 103. When the selection button 803 is pressed with one or more member-side recommended actions 822 selected in the check boxes 823, the decision-making support apparatus 100 saves the member-side recommended actions 822 selected in the check boxes 823 as a recommended action selection result 804. In this example, the recommended action selection result 804 is "active information sharing" (the recommended action ID 501 is R02M).

[0048] [Population Data Input Screen] Figure 9 is an explanatory diagram showing an example of a population data input screen. The population data input screen 900 is displayed by pressing the select button 803. The population data input screen 900 displays the recommended action selection results 804 as selected recommended actions. The population data input screen 900 has a population data input area 901 and an analyze button 902.

[0049] The group data input area 901 has input items related to, for example, the number of people in the group, the average age, the female ratio, educational background, nationality, place of activity, number of years of activity, organizational structure, intimacy, and cooperation.

[0050] The number of people in a group indicates the range to which the number of members of the target group belongs. The decision support device 100 identifies the number of members of the group in the answer data (question score 302) from the member DB 200, identifies the range to which the number of members belongs, and displays it as the number of people in the group. The number of people in a group can also be selected by a user (for example, the leader of the target group) by operating the input device 103. In this example, the number of members in the group is 35, so the number of means is displayed as "30 to 50."

[0051] The average age is the average value of the ages 202 of the members of the target group. The decision support device 100 identifies the ages 202 of the members of the group from the member DB 200 in the answer data (question score 302), calculates the average value of the ages 202, and displays it as the average age.

[0052] The female ratio is the proportion of the target group whose gender 203 is female. The decision support device 100 identifies the number of group members whose gender 203 is female in the answer data (question score 302) from the member DB 200, and divides it by the total number of members of the group to calculate and display the female ratio.

[0053] The educational background is a rough educational background of the target group. The group size is selected by the user (for example, the leader of the target group) by operating the input device 103.

[0054] Nationality is a qualification indicating that members of a target group belong to a specific country and are its citizens. It indicates the distribution of nationalities across the group as a whole, rather than the nationalities 204 of individual members. The decision support device 100 may read out the nationality 204 of each member and select from a number of categories, such as "Japanese only" or "including foreigners," or a user (e.g., the leader of the target group) may select one by operating the input device 103.

[0055] The location of activity, number of years of activity, organizational structure, intimacy, and cooperativeness are selected by the user (for example, the leader of the target group) by operating the input device 103.

[0056] The analysis button 902 is a user interface that can be pressed by a user (for example, the leader of the target group) by operating the input device 103. When the analysis button 902 is pressed, the input data (group size, average age, female ratio, educational background, nationality, location of activity, years of activity, organizational structure, intimacy, and cooperativeness) entered in the group data input area 901 is saved as group data 910.

[0057] The recommended action analysis process (step S604) can also be realized by machine learning using the group type 403 as a feature and the recommended actions 502 as correct answer data, rather than the rule-based method described above.

[0058] 6 , in the promotion measure analysis process (step S605), the decision support apparatus 100 analyzes promotion measures. The promotion measures are measures that the target group should take to promote the selected recommended behavior of the recommended behavior selection result 804.

[0059] 10 is an explanatory diagram showing an example of the promotion measure analysis process (step S605). In the promotion measure analysis process (step S605), the decision support apparatus 100 converts the recommended action selection result 804 and the population data 910 into data.

[0060] In the promotion policy analysis process (step S605), the decision-making support device 100 inputs the population data 910 and the recommended action selection result 804, performs data conversion (step S1001), and outputs a reference feature 1011. The decision-making support device 100 also performs addition (step S1002) of the reference feature 1011 and the policy feature table 1012. The decision-making support device 100 then inputs the addition result of step S1002 to the machine learning model 1013 to perform cooperation rate prediction (step S1003). After this, the decision-making support device 100 aggregates the cooperation rates, which are the prediction results of the cooperation rate prediction (step S1003) (step S1004). The data conversion (step S1001), addition (step S1002), cooperation rate prediction (step S1003), and result aggregation (step S1004) will be described in order below.

[0061] 11 is an explanatory diagram showing an example of a first data conversion table in data conversion (step S1001). The first data conversion table 1100 is a table for converting the population data 910 into feature quantities (populations) F51 to F75. The first data conversion table 1100 specifies a data item to be converted 1101, a feature quantity (population) 1102, and a conversion algorithm 1103.

[0062] The data items 1101 to be converted are the data items in the group data 910 that are the target of data conversion in the data conversion (step S1001) (group size, average age, female ratio, educational background, nationality, location of activity, years of activity, organizational structure, intimacy, cooperativeness).

[0063] The feature (group) 1102 is a feature related to a group, and serves as an index for the conversion target data item 1101 in the same row. For example, F51 means a feature related to the number of people in a group. In FIG. 11, there are 25 types of feature (group) 1102 (F51 to F25), but the number of types is not limited to 25.

[0064] The conversion algorithm 1103 is a function that converts the value of the conversion target data item 1101 into a feature (group) 1102. For example, in the case of F51, the set range for the group size is "30 to 50," so the "center of the set range" is "40." Therefore, F51 is calculated using the following formula (1).

[0065] F51 = log (center of setting range) ÷ log (1000) = log (40) ÷ log (1000) = 0.534 (1)

[0066] Furthermore, for the feature quantities F72 to F75 (agreement), one of the feature quantities F72 to F75 is set to "1" and the rest are set to "0." In the example of Figure 9, because agreeableness is set to "individualism," F72 is set to "1" and F73 to F75 are set to "0." In this way, the decision-making support device 100 acquires the feature quantities F51 to F75 related to the group from the group data 910.

[0067] 12 is an explanatory diagram showing an example of a second data conversion table in the data conversion (step S1001). The second data conversion table 1200 is a table for converting the recommended action selection result 804 into a feature amount (situation).

[0068] The second data conversion table 1200 has fields of a selected recommended action ID 1201 and feature quantities (recommended actions) 1202 (F26 to F50). The selected recommended action ID 1201 is identified by the recommended action selection result 804. The feature quantities (recommended actions) 1202 are feature quantities F26 to F50 related to the recommended action. The feature quantities F26 to F50 take different values ​​for each feature quantity (recommended action) 1202. In this example, the recommended action selection result 804 has the value "R02M" of the selected recommended action ID 1201 (see FIGS. 8 and 9). Therefore, the decision support device 100 acquires the feature quantities F26 to F50 of the row of "R02M".

[0069] In FIG. 12, there are 25 types (F26 to F50) of feature amounts (recommended actions) 1202, but the number of types is not limited to 25.

[0070] 13 is an explanatory diagram showing an example of the reference feature 1011 output by the data conversion (step S1001). The reference feature 1011 has the following fields: a reference ID 1300, a feature (measure) 1301, a feature (recommended action) 1202, and a feature (population) 1102.

[0071] The reference ID 1300 is identification information that uniquely identifies the reference feature 1011. The feature (measure) 1301 is feature F01 to F25 related to the measure. The feature F01 to F25 related to the measure is set for each measure, but the reference feature 1011 is set to 0 because it serves as the reference for all measures.

[0072] The feature amounts F26 to F50 converted using the second data conversion table 1200 in Fig. 12 are set in the feature amount (recommended action) 1202. The feature amounts F51 to F75 converted using the first data conversion table 1100 in Fig. 11 are set in the feature amount (population) 1202.

[0073] 14 is an explanatory diagram showing an example of the policy feature table 1012. The policy feature table 1012 is a table that specifies, for each policy, a feature related to the policy to be taken by the group. The policy feature table 1012 has a policy ID 1400, a feature (policy) 1401, a feature (recommended action) 1402, and a feature (group) 1403.

[0074] The policy ID 1400 is identification information that uniquely identifies a policy. In this example, 25 types of policies (T01 to T25) are defined. In FIG. 14, there are 25 types of policy IDs 1400 (T01 to T25), but the number of types is not limited to 25.

[0075] The feature (measure) 1401 is a feature related to a measure that uniquely characterizes the measure identified by the measure ID 1400. In this example, a feature whose last digit in the measure ID 1400 matches a feature whose last digit in the feature (measure) 1401 matches is set to "1," and a feature whose last digit does not match is set to "0." For example, if the measure ID 1400 is "T01," the value of the feature F01 is set to "1," and the values ​​of the feature F2 to F25 are set to "0." In other words, the feature (measure) 1401 is expressed as a diagonal matrix whose diagonal elements are "1" when viewed as a whole across measures T01 to T25. Note that, although a diagonal matrix is ​​used in this example, it need not be a diagonal matrix as long as the values ​​of the feature F1 to F25 differ for each measure.

[0076] In the measure feature table 1012, to express the feature (measure) 1401, the feature (recommended action) 1402 and the feature (population) 1403 are set to "0".

[0077] 15 is an explanatory diagram showing an example of addition (step S1002). In addition (step S1002), the reference feature 1011 is added to the policy feature table 1012. Specifically, for example, the feature (policy) 1301 is added to the feature (policy) 1401, the feature (recommended action) 1202 is added to the feature (recommended action) 1402, and the feature (population) 1102 is added to the feature (population) 1403. As a result, the decision support apparatus 100 obtains an addition result 1500.

[0078] The addition result 1500 has a feature amount (measure) 1401, a feature amount (recommended action) 1402, and a feature amount (population) 1403. That is, the addition result 1500 shows feature amounts F1 to F75 in which the feature amount (recommended action) 1402 and the feature amount (population) 1403 are taken into consideration in the measure.

[0079] Cooperation Rate Prediction (Step S1003) and Result Aggregation (Step S1004) FIG. 16 is an explanatory diagram showing an example of cooperation rate prediction (step S1003). In cooperation rate prediction (step S1003), the addition result 1500 and the reference feature 1011 are each input to the machine learning model 1013. The machine learning model 1013 outputs a first prediction result 1601 when the addition result 1500 is input, and outputs a second prediction result 1602 when the reference feature 1011 is input. The first prediction result 1601 is a set of cooperation rates 1613 for each measure. The second prediction result 1602 is the cooperation rate 1613 when the measure is not applied.

[0080] The cooperation rate 1613 is the probability of predicting whether members of the target group will cooperate to perform the selected recommended behavior if the target group implements the measure (or if the measure is not implemented in the second prediction result 1602). For example, if no measures are implemented in the target group, the cooperation rate 1613 of members of the target group cooperating to perform the selected recommended behavior #R02M "active information sharing" is 54.8%. On the other hand, if the promotion measure 1612 ranked first in the ranking 1611, measure #T02 "discuss the meaning of taking the recommended behavior," is implemented in the target group, the cooperation rate 1613 of members of the target group cooperating to perform the selected recommended behavior #R02M "active information sharing" is 70.3%.

[0081] The machine learning model 1013 is a model that has been trained in advance using a training data set that is composed of sample data (which may be just the reference feature 1011) equivalent to the addition result 1500 and its correct answer data. The correct answer data may be a label indicating "1" when cooperation occurs and "0" when no cooperation occurs, or may be a probability value between 0 and 1.

[0082] The decision support device 100 retrains the machine learning model 1013 by backpropagating errors using the value of a loss function based on the difference between the cooperation rate 1613 indicated by the first prediction result 1601 and the second prediction result 1602 and the actual cooperation rate when the policy is implemented (which may be the subjective opinion of the leader of the group).

[0083] In addition, in the result aggregation (step S1004), the decision-making support apparatus 100 displays a promotion measure analysis result screen 1600 based on the first prediction result 1601 and the second prediction result 1602. The promotion measure analysis result screen 1600 includes a recommended action selection result 804 and a promotion measure analysis result 1610.

[0084] The promotion measure analysis result 1610 has fields of rank 1611, promotion measure 1612, cooperation rate 1613, and improvement 1614. The rank 1611 indicates the rank of the promotion measure 1612 sorted in descending order of the cooperation rate 1613 in the second prediction result 1602. The higher the cooperation rate 1613, the higher the ranking. Note that no measure is set, i.e., for the second prediction result 1602, the rank 1611 is not set.

[0085] The promotion measure 1612 is a character string associated with the measure ID 1400. Note that, for no measure, i.e., the second prediction result 1602, "no measure (B00)" is displayed. "B00" is the reference ID 1300.

[0086] The cooperation rate 1613 is the first prediction result 1601 and the second prediction result 1602 output from the machine learning model 1013. The improvement 1614 is the absolute value of the difference obtained by subtracting the cooperation rate 1613 (without measures) of the second prediction result 1602 from the cooperation rate 1613 assigned the rank 1611 of the first prediction result 1601. If the difference is positive, the absolute value of the difference and an upward arrow are displayed, and if the difference is negative, the absolute value of the difference and an upward arrow are displayed.

[0087] For example, if the promotion measure 1612 of measure #T02, which has the highest ranking 1611, "discuss the meaning of the recommended behavior," is applied to the group, the cooperation rate 1613 (predicted value) is predicted to be 70.3%, which is a 15.5% increase compared to the cooperation rate 1613 (54.8%) without the measure.

[0088] 6 , the decision-making support device 100 executes the policy implementation process (step S606). In the policy implementation process (step S606), the decision-making support device 100 receives input of the results of implementing the promotion policy through an operation by a user (e.g., a leader of a target group), and stores the input data in the storage device 102.

[0089] 17 is an explanatory diagram showing an example of a promotion measure implementation result input screen. The promotion measure implementation result input screen 1700 has a characteristic ID 401, a promotion measure implementation result 1701, and a confirm button 1710. The characteristic ID 401 indicates the characteristic ID of the target group. In this example, it is G61.

[0090] The promotion measure implementation result 1701 includes a recommended action 1702, a promotion measure 1703, and an implementation result 1704. The recommended action 1702, the promotion measure 1703, and the implementation result 1705 are displayed separately for the leader side (L) and the member side (M). The recommended action 1702, the promotion measure 1703, and the implementation result 1704 for the leader side (L) are suffixed with "L," and the recommended action 1702, the promotion measure 1703, and the implementation result 1704 for the member side (M) are suffixed with "M."

[0091] A recommended action 1702L shows the recommended action 821 for the leader side corresponding to the member side recommended action selection result 804 selected in Fig. 8. In this example, the recommended action #R02L "conversation with the member side" is displayed as the recommended action 1702L.

[0092] A recommended action 1702L shows recommended action #R02M, which is the member-side recommended action selection result 804 selected in Fig. 8. In this example, "active information sharing" of recommended action #R02M is displayed as recommended action 1702M.

[0093] The promotion measure 1703L is not displayed because in this example, the group data 910 is generated based on the member's recommended action selection result 804, and not based on the recommended action 1702L.

[0094] The promotion measure 1703M indicates the promotion measure 1612 selected from the promotion measure analysis result 1610. The promotion measure 1703M is, for example, a pull-down user interface that allows a user (e.g., a leader of a target group) to select the promotion measure 1612 by operating the input device 103. In the example of FIG. 17 , measure #T02, "Discuss the meaning of the recommended behavior," is selected. Note that the selection items for the promotion measure 1703M include "Not implemented" in addition to the promotion measure 1612.

[0095] The implementation result 1704L indicates whether or not the promotion measure has been implemented. The implementation result 1704L is a pull-down user interface that allows a user (for example, a leader of a target group) to select "not implemented" or "implemented" by operating the input device 103. In the example of FIG. 17, "implemented" is selected.

[0096] The implementation result 1704M includes a pre-implementation cooperation rate 1741M and a post-implementation cooperation rate 1742M. The pre-implementation cooperation rate 1741M is the cooperation rate before the promotion measure 1612 selected in the promotion measure 1703M is implemented, and is input information input by a user (for example, a leader of the target group) by operating the input device 103.

[0097] The post-implementation cooperation rate 1742M is the cooperation rate after the promotion measure 1612 selected in the promotion measure 1703M is implemented, and is input information input by a user (for example, a leader of the target group) by operating the input device 103. Note that if the promotion measure 1703M is "not implemented," the post-implementation cooperation rate 1742M is also automatically set to "not implemented."

[0098] The Confirm button 1710 is a user interface that can be pressed by a user (for example, the leader of the target group) by operating the input device 103. When the Confirm button 1710 is pressed, the decision-making support apparatus 100 saves a promotion measure implementation result 1720 in the storage device 102. The promotion measure implementation result 1720 includes the input data in the promotion measure implementation result 1701 (promotion measure 1703M, pre-implementation cooperation rate 1741M, post-implementation cooperation rate 1742M), the addition result 1500, and the reference feature 1011. This completes the policy implementation process (step S606).

[0099] 6 , the decision-making support apparatus 100 receives an input indicating that improvement effects have been confirmed by a user (e.g., the leader of the target group) operating the input device 103 (step S607). If the input indicating that improvement effects have been confirmed is received (step S607: Yes), the process returns to step S601. On the other hand, if the input indicating that improvement effects have been confirmed is not received and the process ends (step S607: No), the process ends.

[0100] If the group type 403 of the target group is known, the decision support apparatus 100 may start execution from the recommended behavior analysis process (step S604).

[0101] As described above, according to the first embodiment, it is possible to automatically present the cooperation rate 1613 among the members constituting the small group when the promotion measure 1612 is implemented on the target group. This also increases the possibility that the selected recommended action indicated by the recommended action selection result 804 will be successful when the promotion measure 1612 is implemented on the target group, thereby improving the characteristics of the target group.

[0102] Next, a second embodiment will be described. In the second embodiment, a decision support system that manages a plurality of decision support devices 100 according to the first embodiment will be described. In the second embodiment, differences from the first embodiment will be mainly described, and therefore, parts common to the first embodiment will be assigned the same reference numerals and descriptions thereof will be omitted.

[0103] 18 is an explanatory diagram showing an example of an information processing system. The information processing system includes a management device 1801 and multiple decision-making support devices 100. The management device 1801 and the multiple decision-making support devices 100 are communicably connected via a network 1802 such as the Internet, a local area network (LAN), or a wide area network (WAN). The management device 1801 is a decision-making support device realized by, for example, the hardware configuration shown in FIG. 1. The management device 1801 may also be any one of the multiple decision-making support devices 100.

[0104] The management device 1801 has a survey implementation result group 1811, a recommended action master table 1812, a promotion measure implementation result group 1813, and a machine learning master model 1814. Specifically, the survey implementation result group 1811, the recommended action master table 1812, the promotion measure implementation result group 1813, and the machine learning master model 1814 are stored, for example, in the storage device 102 of the management device 1801, but may also be stored in another computer accessible to the management device 1801 via the network 1802 by the communication IF 105 of the management device 1801.

[0105] The survey implementation result group 1811 will be described later with reference to Fig. 19. The promotion measure implementation result group 1813 will be described later with reference to Fig. 23. The recommended action master table 1812 is a table with the same structure as the recommended action table 500. The machine learning master model 1814 is a model with the same structure as the machine learning model 1013.

[0106] The recommended action master table 1812 and the machine learning master model 1814 are updated by the management device 1801. The management device 1801 distributes the recommended action master data in the latest updated recommended action master table 1812 and the learning parameters constituting the machine learning master model 1814 to each decision-making support device 100. The decision-making support device 100 receives the recommended action master data in the recommended action master table 1812 and the learning parameters constituting the machine learning master model 1814 from the management device 1801, and updates the recommended action table 500 and the machine learning model 1013, which serve as slave information.

[0107] <Updating the Recommended Action Table 500> Next, an example of updating the recommended action table 500 will be described.

[0108] Fig. 19 is an explanatory diagram showing an example of the survey implementation result group 1811 shown in Fig. 18. The survey implementation result group 1811 has survey implementation results for each of the multiple decision-making support devices 100. The survey implementation result group 1811 has, as fields, a computer ID 1900, a characteristic ID 401, a selected recommended behavior ID 1201, and a total score 303.

[0109] The computer ID 1811 is identification information that uniquely identifies a computer, that is, the decision support device 100. Therefore, one entry indicates one survey result. For example, entry 1911 is the survey result obtained by the decision support device 100 of computer #C001.

[0110] The total score 303 has subfields, a pre-measure 1901 and a post-measure 1902. The pre-measure 1901 is the total score 303 for a questionnaire administered to a target group before the promotion measure 1612 corresponding to the selected recommended behavior ID 1201 is administered to the target group in the decision support device 100 identified by the computer ID 1811 (see FIG. 3 ).

[0111] Post-measure 1902 is the total score 303 for a questionnaire (the same questionnaire as pre-measure 1901) administered to a target group after the promotion measure 1612 corresponding to the selected recommended behavior ID 1201 was implemented in the target group in the decision support device 100 identified by the computer ID 1811 (see Figure 3).

[0112] For example, entry 1911 indicates that in the decision support device 100 of computer #C001, when a target group having characteristic #G61 corresponds to selected recommended action #R02M, the total score 303 of the responses to a questionnaire administered before the measure 1901 is "3.78," and the total score 303 of the responses to the same questionnaire administered after the measure 1902 is "3.99." Note that if a questionnaire has not been administered after the measure 1902, "not administered" is stored in after the measure 1902, as shown in entry 1912.

[0113] 20 is an explanatory diagram showing an example of update data 2000. The update data 2000 is data generated by the management device 1801 with reference to the survey implementation result group 1811. The update data 2000 has fields: a correction target 2001 and a correction value 2002. The correction target 2001 is a combination of the characteristic ID 401 and the selected recommended action ID 1201 to be corrected by the correction value 2002. For example, in the case of entry 1911 in FIG. 19 , the characteristic ID 401 is "G61" and the selected recommended action ID 1201 is "R02M," so the characteristic ID 401 and the selected recommended action ID 1201 are concatenated to become "G61R02M."

[0114] The group characteristic ranking 503 of the selected recommended action ID 1201 (recommended action ID 501) in the group type 403 of the target group identified by the characteristic ID 401 is identified as the correction target 2001. For example, in the case of entry 1911 in Fig. 19 , the characteristic ID 401 is "G61" and the selected recommended action ID 1201 is "R02M." Therefore, "1" is identified as the group characteristic ranking 503 in Fig. 5 .

[0115] The correction value 2002 is data for correcting the group characteristic ranking 503 specified by the correction target 2001. The correction value 2002 is composed of a positive or negative sign and an increment width. The sign is determined by the result of comparing the overall score 303 before the measure 1901 with the overall score 303 after the measure 1902. If before the measure 1901 < after the measure 1902, the sign is determined to be negative. If before the measure 1901 > after the measure 1902, the sign is determined to be positive. If before the measure 1901 = after the measure 1902, or if after the measure 1902 has not been implemented, the increment width is determined to be 0.

[0116] The step size is a preset value, and is set to "0.1" in the example of Fig. 20. The step size is not limited to "0.1". The value of the step size may be different for each correction target 2001.

[0117] 21 is an explanatory diagram showing an example of the recommended action master table 1812. For example, when the correction target 2001 is "G61R02M", the correction value 2002 is "-0.1". In the recommended action master table 1812, the value "1" of the group characteristic ranking 503 is identified where the column in which the characteristic ID 401 is "G61" intersects with the row in which the recommended action ID 501 is "R02M".

[0118] The management device 1801 adds "-0.1", which is the correction value 2002 when the correction target 2001 is "G61R02M", to the specified value "1" to calculate "0.9". This "0.9" becomes the value of the specified group characteristic ranking 503 in the recommended action master table 1812. The management device 1801 executes this process for each survey result.

[0119] 22 is a sequence diagram of updating the recommended action table 500. Each of the decision-making support devices 100 transmits the survey results to the management device 1801 (step S2201). The management device 1801 receives the survey results from each of the decision-making support devices 100 and stores them as a survey result group 1811 (step S2202).

[0120] The management device 1801 compares the total score 303 before the action 1901 and after the action 1902 for each survey result, calculates a correction value 2002, and generates update data 2000 (step S2203).

[0121] The management device 1801 corrects the ranking of the correction target 2001 for each survey result as shown in FIG. 21 based on the update data 2000 in step S2203, and updates the recommended action master table 1812 (step S2204).

[0122] The management device 1801 distributes the recommended action master data in the recommended action master table 1812 after the update in step S2204 to each decision support device 100 (step S2205). The recommended action master data is matrix data of the group characteristic ranking 503 in the recommended action master table 1812 after the update in step S2204. Note that the distributed recommended action master data may be the correction target 2001 and the updated value of the group characteristic ranking 503.

[0123] Each decision-making support device 100 updates the recommended action table 500 with the received recommended action master data (step S2206). As a result, in each decision-making support device 100, the recommended action table 500 coincides with the recommended action master table 1812. This completes the update of the recommended action table 500.

[0124] <Updating the Machine Learning Model 1013> Next, an example of updating the machine learning model 1013 will be described.

[0125] Fig. 23 is an explanatory diagram showing an example of the promotion measure implementation result group 1813 shown in Fig. 18. The promotion measure implementation result group 1813 has, as fields, a computer ID 1900, a promotion measure ID 2301, a feature amount 2302, and a cooperation rate result 2303. Two rows of entries having the same value for the computer ID 1900 correspond to the promotion measure implementation result 1720 from the decision support device 100 identified by that computer ID 1900.

[0126] The promotion measure ID 2301 includes the reference ID 1300 (no measure) and the measure ID 1400 of the promotion measure 1703M.

[0127] The feature quantity 2302 includes the values ​​of the feature quantities F01 to F75 of the reference ID 1300 (no measure) and the values ​​of the feature quantities F01 to F75 of the promotion measure 1703M. The values ​​of the feature quantities F01 to F75 of the reference ID 1300 (no measure) are, for example, the reference feature quantity 1011. The values ​​of the feature quantities F01 to F75 of the promotion measure 1703M are, for example, the sum result 1500.

[0128] The cooperation rate result 2303 includes a pre-implementation cooperation rate 1741M for the reference ID 1300 (no measure) and a post-implementation cooperation rate 1742M for the promotion measure 1703M. Note that in the cooperation rate result 2303, "not input" indicates that the implementation result 1704M has not yet been input in the decision support device 100 identified by the computer ID 1900.

[0129] The two rows of entries with the same value for computer ID 1900 are promotion measure implementation results 1720, which are information regarding the learning of the machine learning master model 1814 in the machine learning model 1013 (hereinafter referred to as learning information), i.e., information used for relearning the machine learning master model 1814.

[0130] 24 is a sequence diagram of updating the machine learning model 1013. Each of the multiple decision-making support devices 100 transmits learning information to the management device 1801 (step S2401). Note that if each decision-making support device 100 has completed re-learning using the promotion measure implementation result 1720, each decision-making support device 100 may transmit the learning parameters of the machine learning model 1013 as learning information.

[0131] The management device 1801 receives the promotion measure implementation results 1720 from each of the decision support devices 100 and stores them as a promotion measure implementation result group 1813 (step S2402).

[0132] The management device 1801 retrains the machine learning master model 1814 using the features F01 to F75 of the promotional measure implementation result group 1813 (step S2403). Specifically, for example, the management device 1801 uses the features 2302 as training data and the cooperation rate result 2303 as correct data, inputs the training data into the machine learning master model 1814, and predicts the cooperation rate. The management device 1801 retrains the machine learning master model 1814 by backpropagation using the value of a loss function based on the difference between the predicted cooperation rate and the correct data. Note that the management device 1801 does not use promotional measure implementation results 1720 for which the cooperation rate result 2303 is "not input" for retraining.

[0133] Furthermore, the management device 1801 may execute re-learning (step S2403) when a predetermined number or more of promotion measure implementation results 1720 for which the cooperation rate result 2303 is not "not input" have been accumulated. This eliminates the need to wait for re-learning (step S2403) until the "not input" state of the cooperation rate result 2303 is resolved, thereby shortening the update interval of the machine learning model 1013.

[0134] In addition, if the learning information is a learning parameter of the machine learning model 1013, the management device 1801 may average the learning parameters from each decision support device 100 through federated learning and update the machine learning master model 1814 using the averaged learning parameters.

[0135] Then, the management device 1801 distributes the learning parameters of the updated machine learning master model 1814 to each decision support device 100 (step S2404).

[0136] Each decision support device 100 updates the machine learning model 1013 with the learning parameters from the management device 1801 (step S2106). As a result, in each decision support device 100, the machine learning model 1013 coincides with the machine learning master model 1814. This completes the update of the machine learning model 1013.

[0137] Thus, according to the second embodiment, the management device 1801 manages multiple decision-making support devices 100 in an integrated manner, thereby reducing the variation in the likelihood that the selected recommended action indicated by the recommended action selection result 804 will be successful among the multiple decision-making support devices 100, and improving the characteristics of the target group in each of the multiple decision-making support devices 100.

[0138] The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added to, deleted from, or replaced with other configurations.

[0139] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by having a processor interpret and execute a program that realizes each function.

[0140] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, hard disk, or SSD (Solid State Drive), or in a recording medium such as an IC (Integrated Circuit) card, SD card, or DVD (Digital Versatile Disc).

[0141] In addition, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily represent all the control lines and information lines that are necessary for implementation. In reality, it can be assumed that almost all components are interconnected.

Claims

1. A decision support device having a processor that executes a program and a storage device that stores the program, The processor: an acquisition process for acquiring a feature amount related to a measure to be taken by a group, a feature amount related to the behavior of the group, and a feature amount related to the group; a first prediction process in which, when the feature amounts related to the behavior and the feature amounts related to the group are input according to whether or not the measure is implemented, the feature amounts related to the behavior, and the feature amounts related to the group acquired by the acquisition process are input into a machine learning model that predicts whether members of the group will cooperate to take the behavior, and outputs a first prediction result that predicts the behavior of members of the group in cooperation; an output process for outputting a first prediction result obtained by the first prediction process; A decision support device that executes the above.

2. 2. The decision support device according to claim 1, The processor: a second prediction process that inputs the feature amount related to the behavior and the feature amount related to the group into the machine learning model, and outputs a second prediction result in which members of the group cooperate to predict the behavior; In the output process, the processor outputs a second prediction result obtained by the second prediction process. A decision support device characterized by:

3. 2. The decision support device according to claim 1, In the acquisition process, the processor accepts a selection of a specific behavior from one or more recommended behaviors, and acquires a feature amount related to the specific behavior of the group; In the first prediction process, the processor inputs the feature amount related to the measure, the feature amount related to the specific behavior, and the feature amount related to the group into the machine learning model, and outputs a first prediction result in which members of the group cooperate to predict the specific behavior. A decision support device characterized by:

4. 2. The decision support device according to claim 1, The processor: performing a recommended behavior analysis process that identifies one or more recommended behaviors based on characteristics of the population; In the acquisition process, the processor accepts a selection of a specific behavior from the one or more recommended behaviors identified by the recommended behavior analysis process, and acquires a feature amount related to the specific behavior of the group. A decision support device characterized by:

5. 5. The decision support device according to claim 4, The processor: execute a group characteristic analysis process for analyzing the characteristics of the group based on the answers to the questions posed to each of the members; In the recommended behavior analysis process, the processor identifies the one or more recommended behaviors based on the characteristics of the group analyzed by the group characteristic analysis process. A decision support device characterized by:

6. 2. The decision support device according to claim 1, The processor: a learning process for learning the machine learning model based on the first prediction result and input information indicating the result of the members cooperating to take the action when the measure is implemented; A decision support device that executes the above.

7. A decision support device that has a processor that executes a program and a storage device that stores the program, and is capable of communicating with a plurality of computers, the decision support device holds master information in which recommended actions for the group are ranked according to characteristics of the group; each of the plurality of computers holds slave information in which a ranking of recommended actions for the group is set according to characteristics of the group; The processor: a receiving process for receiving from each of the plurality of computers an implementation result including a specific characteristic of the group, a specific recommended action for the group, a first score for the group based on a first answer to a question given to each of the members constituting the group before the implementation of the measure to be taken by the group, and a second score for the group based on a second answer to the question given to each of the members after the implementation of the measure; Based on the first score and the second score received by the receiving process, A calculation process for calculating a correction value for a ranking of a specific recommended action for the group based on a specific characteristic of the group; a correction process for correcting the ranking of the specific recommended action based on the correction value calculated by the calculation process; a transmission process of transmitting a correction result obtained by the correction process to the plurality of computers; A decision support device that executes the above.

8. A decision support device that has a processor that executes a program and a storage device that stores the program, and is capable of communicating with a plurality of computers, the decision support device holds master information in which recommended actions for the group are ranked according to characteristics of the group; each of the plurality of computers holds a machine learning master model that, when a feature related to the behavior of the group and a feature related to the group are input, predicts whether members of the group will cooperate to take the action, depending on whether a measure to be taken by the group is implemented; each of the plurality of computers holds a machine learning slave model that, when a feature related to the behavior and a feature related to the group are input, predicts whether members of the group will cooperate to take the behavior, depending on whether the measure is implemented; The processor: a receiving process for receiving information regarding the learning of the machine learning master model in the machine learning slave model from each of the plurality of computers; a learning process for learning the machine-learning master model based on information related to the learning received from each of the plurality of computers by the receiving process; a transmission process of transmitting learning parameters of the machine learning master model learned by the learning process to the plurality of computers; A decision support device that executes the above.

9. 9. The decision support device according to claim 8, the information related to the learning includes a feature related to the measure, a feature related to the behavior, a feature related to the group, and first input information indicating a result of the members cooperating to take the action when the measure is implemented; In the learning process, the processor learns the machine-learning master model based on a first prediction result output as a result of inputting the feature related to the policy, the feature related to the behavior, and the feature related to the group, which are received from each of the plurality of computers in the receiving process, into the machine-learning master model, and based on the first input information. A decision support device characterized by:

10. 10. The decision support device according to claim 9, the information about the learning includes second input information indicating a result of the members cooperating to take the action when the measure is not implemented, In the learning process, the processor learns the machine learning master model based on a second prediction result output as a result of inputting the feature amount related to the behavior and the feature amount related to the group into the machine learning master model, and based on the second input information. A decision support device characterized by:

11. 9. The decision support device according to claim 8, the information about the learning is a learning parameter of the machine learning slave model; In the learning process, the processor learns the machine-learning master model based on the learning parameters received from each of the plurality of computers in the receiving process. A decision support device characterized by:

12. 9. The decision support device according to claim 8, The processor: executing the learning process when information related to the learning is received from a predetermined number or more of the plurality of computers; A decision support device characterized by:

13. A decision support method executed by a decision support apparatus having a processor that executes a program and a storage device that stores the program, comprising: The processor: an acquisition process for acquiring a feature amount related to a measure to be taken by a group, a feature amount related to the behavior of the group, and a feature amount related to the group; a first prediction process in which, when the feature amounts related to the behavior and the feature amounts related to the group are input according to whether or not the measure is implemented, the feature amounts related to the behavior, and the feature amounts related to the group acquired by the acquisition process are input into a machine learning model that predicts whether members of the group will cooperate to take the behavior, and outputs a first prediction result that predicts the behavior of members of the group in cooperation; an output process for outputting a first prediction result obtained by the first prediction process; A decision support method comprising:

14. The processor an acquisition process for acquiring a feature amount related to a measure to be taken by a group, a feature amount related to the behavior of the group, and a feature amount related to the group; a first prediction process in which, when the feature amounts related to the behavior and the feature amounts related to the group are input according to whether or not the measure is implemented, the feature amounts related to the behavior, and the feature amounts related to the group acquired by the acquisition process are input into a machine learning model that predicts whether members of the group will cooperate to take the behavior, and outputs a first prediction result that predicts the behavior of members of the group in cooperation; an output process for outputting a first prediction result obtained by the first prediction process; A decision support program characterized by executing the above.

Images