Consensus building assistance device, consensus building assistance method, and consensus building assistance program

Abstract

A consensus building assistance device capable of accessing a database in which are stored an option group to be adopted for a measure and a factor group for establishing each option of the option group executes: an acquisition process for acquiring, for each participant, an evaluation result obtained by evaluating each factor of the factor group in a plurality of stages; a classification process for classifying the factor group into a plurality of groups, including a first group to which a positively evaluated first factor belongs and a second group to which a negatively evaluated second factor belongs, on the basis of the evaluation result for each participant acquired using the acquisition process; and an output process for outputting a classification result produced in the classification process.

Description

Consensus building support device, consensus building support method, and consensus building support program

[0001] The present invention relates to a consensus building support device, a consensus building support method, and a consensus building support program that support consensus building.

[0002] When reaching consensus on social issues, such as policy decisions in local governments or communities, it is desirable to select one option from multiple options that all participants agree on. However, differences in values ​​among participants often lead to disagreements, making it difficult to select a single option that everyone agrees on.

[0003] The search device in Patent Document 1 below identifies, for each inquiry, from a plurality of options presented based on the respondent's past response history, a suitable option that corresponds to the respondent's response tendency. When the responses to the plurality of options are included in the suitable options, the search device determines that the mode is a preference attainment mode, in which estimates of the respondent's preferences have converged. When the search device determines that the mode is a preference attainment mode, it changes the options presented compared to when the mode is not determined to be a preference attainment mode.

[0004] Non-Patent Document 1 below discloses a framework for the graph model for conflict resolution (GMCR).

[0005] Japanese Patent Application Publication No. 2019-101655

[0006] Takehiro Ihara, "GMCR: The Graph Model for Conflict Resolution," Operations Research Society of Japan, Vol. 58, No. 4, 2013.

[0007] However, the above-mentioned Patent Document 1 and Non-Patent Document 1 do not disclose a method for generating new options from the presented options, and do not take into consideration the issue of finding options that are acceptable to all participants.

[0008] The present invention aims to support the formation of a convincing agreement among participants.

[0009] A consensus building support device that is one aspect of the invention disclosed in this application is a consensus building support device that can access a database that stores a group of options to be taken for a policy and a group of factors for each option in the group of options to be established, and has a processor that executes a program and a storage device that stores the program, and is characterized in that the processor performs the following steps: an acquisition process that acquires, for each participant, evaluation results obtained by evaluating each factor in the group of factors on a multiple-level scale; a classification process that classifies the group of factors into multiple groups based on the evaluation results for each participant acquired by the acquisition process, including a first group to which a first factor that is evaluated positively belongs and a second group to which a second factor that is evaluated negatively belongs; and an output process that outputs the classification results obtained by the classification process.

[0010] According to the exemplary embodiment of the present invention, it is possible to support the formation of a convincing agreement among participants. Problems, configurations, and effects other than those described above will become clear from the following description of the examples.

[0011] FIG. 1 is an explanatory diagram showing an example of the system configuration of a consensus building system. FIG. 2 is a block diagram showing an example of the hardware configuration of the computer shown in FIG. 1. FIG. 3 is a block diagram showing an example of the functional configuration of a consensus building support apparatus. FIG. 4 is a block diagram showing an example of the functional configuration of a personal device. FIG. 5 is a flowchart showing an example of a process procedure for creating presentation data for factor evaluation by the consensus building support apparatus. FIG. 6 is an explanatory diagram showing an example of presentation data for factor evaluation. FIG. 7 is an explanatory diagram showing an example of a display screen of a factor evaluation input unit in a personal device. FIG. 8 is a flowchart showing an example of a process procedure for generating new options by the consensus building support apparatus. FIG. 9 is an explanatory diagram showing an example of the data structure of a factor evaluation data group. FIG. 10 is an explanatory diagram showing an example of group classification by the factor classification unit. FIG. 11 is an explanatory diagram showing an example of an evaluation word group. FIG. 12 is an explanatory diagram showing an example of assignment of evaluation words by the condition generation unit. FIG. 13 is an explanatory diagram showing an example of conditional statement generation by the condition generation unit. FIG. 14 is an explanatory diagram showing an example of generation of new options by the subversion plan generation unit. FIG. 15 is an explanatory diagram showing an example of display of new options on a personal device. FIG. 16 is an explanatory diagram showing another example 1 of score calculation. FIG. 17 is an explanatory diagram showing another example 2 of score calculation. FIG. 18 is a two-dimensional scatter plot with the first principal component and the second principal component in the principal component analysis shown in FIG. 17 as axes. FIG. 19 is an explanatory diagram showing another example 1 of group classification by the factor classification unit. FIG. 20 is an explanatory diagram showing another example 2 of group classification by the factor classification unit. FIG. 21 is an explanatory diagram showing an example of generating a group of factors in a database. FIG. 22 is an explanatory diagram showing another example of generating new options. FIG. 23 is an explanatory diagram showing an example of dividing measures. FIG. 24 is an explanatory diagram showing an example of dividing factors.

[0012] 1 is an explanatory diagram showing an example of the system configuration of a consensus building system. The consensus building system 100 includes a consensus building support apparatus 101 that functions as a server, a database 102, and a personal device 103. The consensus building support apparatus 101, the database 102, and the personal device 103 are communicatively connected via a network 110 such as the Internet, a local area network (LAN), or a wide area network (WAN).

[0013] The consensus building support apparatus 101 is a computer that performs calculations to generate new options for consensus building among participants. The database 102 is a storage device that stores options, their factors, and evaluation words based on evaluation values. The database 102 may be configured inside the consensus building support apparatus 101. The personal device 103 is a computer such as a personal computer, tablet terminal, or smartphone, and transmits and receives input data from users or output data from the consensus building support apparatus 101 via the network 110. The user of the personal device 103 is a participant in a meeting or other target for consensus building.

[0014] In addition, a text generation AI (Artificial Intelligence) 104 is connected to the network 110. The text generation AI 104 is a type of artificial intelligence that applies natural language processing technology, and has a sentence generation function that, upon receiving a question, interprets its structure, grammar, and meaning and returns an answer. The text generation AI 104 is an interactive language model, and is constructed, for example, based on self-supervised learning. The text generation AI 104 may be implemented inside the consensus building support device 101.

[0015] <Example of Computer Hardware Configuration> Figure 2 is a block diagram showing an example of the hardware configuration of the computer (consensus building support apparatus 101, personal device 103) shown in Figure 1. The computer 200 has a processor 201, a storage device 202, an input device 203, an output device 204, and a communication interface (communication IF) 205. The processor 201, the storage device 202, the input device 203, the output device 204, and the communication IF 205 are connected by a bus 206. The processor 201 controls the computer 200. The storage device 202 serves as a working area for the processor 201. The storage device 202 is also a non-temporary or temporary recording medium that stores various programs and data. The storage device 202 may be, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), a HDD (Hard Disk Drive), or a flash memory. The input device 203 inputs data. The input device 203 may be, for example, a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, a microphone, or a sensor. The output device 204 outputs data. The output device 204 may be, for example, a display, a printer, or a speaker. The communication IF 205 connects to the network 110 and transmits and receives data.

[0016] 3 is a block diagram showing an example of the functional configuration of the consensus building support device 101. The consensus building support device 101 has an input unit 301, a factor evaluation presentation data creation unit 302, a factor evaluation result acquisition unit 303, a factor classification unit 304, a condition generation unit 305, a synthesis proposal generation unit 306, an output unit 307, factor evaluation presentation data 308, and a factor evaluation data group 309.

[0017] The input unit 301, the presentation data creation unit for factor evaluation 302, the factor evaluation result acquisition unit 303, the factor classification unit 304, the condition generation unit 305, the synthesis proposal generation unit 306, and the output unit 307 are specifically functions that are realized, for example, by having the processor 201 execute a program stored in the storage device 202 shown in FIG. 2.

[0018] The presentation data for factor evaluation 308 and the factor evaluation data group 309 are data stored in the storage device 202 .

[0019] The database 102 also stores an existing option group 311, an existing option factor group 312 that is each factor of the existing option group 311, and an evaluation word group 313 based on each evaluation value of the factor group 312. The existing option group 311 exists for each measure, and the user of the consensus building support device 101 can select a measure and an existing option group 311 that should be adopted by the measure.

[0020] The input unit 301 accepts data input from the database 102 and the personal device 103. Specifically, for example, the input unit 301 accepts input of an existing option group 311, factor group 312, and evaluation word group 313 from the database 102. The input unit 301 also accepts input of evaluation values ​​for the factor group 312 of each participant from the personal device 103.

[0021] The factor evaluation presentation data creation unit 302 creates factor evaluation presentation data 308 for obtaining each participant's evaluation of the factors, using the existing option group 311 and factor group 312, and transmits the data to the personal device 103 via the output unit 307. The participants are users of the personal device 103.

[0022] The factor evaluation result acquisition unit 303 acquires and tallys the evaluation results for the factors of each participant from the personal device 103. The factor evaluation result acquisition unit 303 stores the tallying results in the storage device 202 as a factor evaluation data group 309.

[0023] The factor classification unit 304 classifies the factor group 312 into a plurality of groups based on the factor evaluation data group 309 .

[0024] The condition generating unit 305 associates the evaluative word group 313 stored in the database 102 based on the classification results of the factor classifying unit 304, and generates a conditional statement for generating a sublation proposal, which is a new option.

[0025] The syntactic suggestion generating unit 306 generates new options using the conditional statements generated by the condition generating unit 305 .

[0026] The output unit 307 outputs the classification results by the factor classification unit 304, the conditional statements generated by the condition generation unit 305, and the syntactic proposals (new options) generated by the syntactic proposal generation unit 306. Specifically, for example, the output unit 307 displays these output data on a display, which is an example of the output device 204, or transmits them to the personal device 103 via the communication IF 205.

[0027] <Example of Functional Configuration of Personal Device 103> Fig. 4 is a block diagram showing an example of the functional configuration of the personal device 103. The personal device 103 has an input unit 401, a factor evaluation presentation data acquisition unit 402, a factor evaluation input unit 403, and an output unit 404. Specifically, the input unit 401, the factor evaluation presentation data acquisition unit 402, the factor evaluation input unit 403, and the output unit 404 are functions that are realized by, for example, having the processor 201 execute a program stored in the storage device 202 shown in Fig. 2.

[0028] The input unit 401 receives the input of the factor evaluation presentation data 308 sent from the consensus building support device 101 .

[0029] The presentation data for factor evaluation acquiring unit 402 presents the presentation data for factor evaluation 308 to the participant, who is the user of the personal device 103. Specifically, for example, the presentation data for factor evaluation acquiring unit 402 displays the presentation data for factor evaluation 308 on a display, which is an example of the output device 204.

[0030] The factor evaluation input unit 403 receives input of the participant's own evaluation of the factors in the factor evaluation presentation data 308 by the participant operating the input device 203 .

[0031] The output unit 404 transmits the participant's own evaluation of the factors input to the factor evaluation presentation data 308 to the consensus building support device 101 .

[0032] 5 is a flowchart showing an example of a processing procedure for creating the presentation data for factor evaluation 308 by the consensus building support device 101. The consensus building support device 101 acquires the existing option group 311 and factor group 312 stored in the database 102 (step S501). The consensus building support device 101 creates the presentation data for factor evaluation 308 using the existing option group 311 and factor group 312 (step S502).

[0033] <Factor Evaluation Presentation Data 308> Fig. 6 is an explanatory diagram showing an example of the factor evaluation presentation data 308. The factor evaluation presentation data 308 has, as fields, existing options 601 and factors 602. The existing options 601 shown in Fig. 6 are options related to employment creation measures, and include, for example, options for attracting companies, supporting entrepreneurship, revitalizing local areas, and developing human resources.

[0034] The factors 602 define each of the group of factors 312, such as "distributing the economy to the region," "taking into consideration local demand," "forming a local community," and "being able to employ necessary personnel," for each of the existing options 601.

[0035] 7 is an explanatory diagram showing an example of a display screen of the factor evaluation input unit 403 in the personal device 103. A display screen 700 includes a factor evaluation 701 of an option, a participant ID 702, and a send button 703.

[0036] The option factor evaluation 701 displays the option factor evaluation 701. The option factor evaluation 701 is display information that combines the presentation data for factor evaluation 308 and an evaluation input area 710.

[0037] Participants can use the input device 203 to input "○", "△", or "×" to indicate their evaluation of the factor 602 in the evaluation input area 710. "○", "△", and "×" respectively indicate a high evaluation (important), a medium evaluation (neither ○ nor ×), and a low evaluation (not important). Note that although the evaluation is here in three stages of "○", "△", and "×", it is not limited to three stages as long as there are multiple stages.

[0038] The participant ID 702 is identification information that uniquely identifies a participant who is a user who owns the personal device 103 .

[0039] The send button 703 is a user interface for sending the evaluation result 711 entered in the evaluation input area 710 and the participant ID 702 to the consensus building support apparatus 101 when pressed on the input device 203 .

[0040] 8 is a flowchart showing an example of a processing procedure for generating a new option by the consensus building support apparatus 101. The consensus building support apparatus 101 acquires, via the input unit 301, the evaluation result 711 of the factor evaluation 701 of the option and the participant ID 702 from each personal device 103 (step S801).

[0041] The consensus building support apparatus 101 creates and stores the factor evaluation data group 309 by using the factor evaluation result acquisition unit 303 to aggregate the evaluation results 711 and the participant IDs 702 acquired in step S801 (step S802).

[0042] The consensus building support apparatus 101 causes the factor classifying unit 304 to analyze the evaluation values ​​of the factor evaluation data group 309 and classify them into a specified number of groups (step S803).

[0043] The consensus building support device 101 assigns the evaluative word group 313 stored in the database 102 to each group of factors 602 by the condition generating unit 305 (step S804).

[0044] The consensus building support apparatus 101 generates conditions for new options using the factors 312 to which the evaluation words 313 have been assigned (step S805).

[0045] The consensus building support apparatus 101 generates options with the conditional statement as an input query using the syntactic proposal generating unit 306 (step S806).

[0046] <Factor Evaluation Data Group 309> Fig. 9 is an explanatory diagram showing an example of the data structure of the factor evaluation data group 309. The factor evaluation data group 309 is information that combines the presentation data for factor evaluation 308 and participant-specific evaluation results 900. The participant-specific evaluation results 900 are evaluation results 711 of the factor evaluations 701 acquired from each personal device 103. In Fig. 9, as an example, the evaluation results 711 of three people whose participant IDs 702 are u1, u2, and u3 are aggregated as the participant-specific evaluation results 900.

[0047] 10 is an explanatory diagram showing an example of group classification by the factor classifying unit 304. The factor classifying unit 304 scores the evaluation result 711 of each participant in the participant-specific evaluation result 900 for each of the factors 602. Specifically, for example, the factor classifying unit 304 converts the three-level evaluation results 711, "○", "△", and "×", into numerical values ​​(for example, △ = "0" and × = "-1"), and tallys them up for each factor 602 to calculate a score 1001.

[0048] The factor classification unit 304 classifies the factors 602 based on the scores 1001. In the example of FIG. 10 , since there are three participants, the theoretical maximum value of the score 1001 is "3," the theoretical minimum value is "-3," and the median value is "0." In FIG. 10 , the factor classification unit 304 calculates the difference between each score 1001 and the theoretical maximum value "3," the theoretical minimum value "-3," and the median value "0." Then, if the difference from the theoretical maximum value "3" is minimum, the factor classification unit 304 sets the classification 1002 of the factor 602 to "A." If the difference from the theoretical minimum value "-3" is minimum, the factor classification unit 304 sets the classification 1002 of the factor 602 to "C." If the difference from the median value "0" is minimum, the factor classification unit 304 sets the classification 1002 of the factor 602 to "B."

[0049] Therefore, if the value of score 1001 is "3" or "2", it is classified as "A", if the value of score 1001 is "1", "0" or "-1", it is classified as "B", and if the value of score 1001 is "-2" or "-3", it is classified as "C".

[0050] <Evaluation Word Group 313> Fig. 11 is an explanatory diagram showing an example of the evaluation word group 313. The evaluation word group 313 is a collection of evaluation words 1100. The evaluation words 1100 are terms assigned to each factor 602 for each existing option 601. In Fig. 11, "desirable" is registered as the evaluation word 1100 that positively evaluates the factor 602, and "undesirable" is registered as the evaluation word 1100 that negatively evaluates the factor 602.

[0051] <Example of Assignment of Evaluation Words 1100> Figure 12 is an explanatory diagram showing an example of assignment of evaluation words 1100 by the factor classifying unit 304. The factor classifying unit 304 assigns evaluation words 1100 stored in the database 102 to the classes 1002. In the example of Figure 12, the factor classifying unit 304 assigns "desirable" as the evaluation word 1100 to the group "A" of the class 1002 with the highest evaluation, and assigns "undesirable" as the evaluation word 1100 to the group "C" of the class 1002 with the lowest evaluation. The factors 602 of the class 1002 to which "desirable" is assigned are referred to as positive factors, and the factors 602 of the class 1002 to which "undesirable" is assigned are referred to as negative factors. Furthermore, the evaluation word 1100 is not assigned to the group "B" with an intermediate evaluation.

[0052] 13 is an explanatory diagram showing an example of a conditional statement generated by the condition generating unit 305. The condition generating unit 305 generates a conditional statement for generating new options according to the classification 1002 of the factor 602 and the assigned evaluation word 1100. Specifically, for example, as an employment creation measure, a conditional statement 1300 is generated in which the classification 1002 assigned the evaluation word 1100 "desirable" satisfies the factor 602 belonging to the "A" group, and the classification 1002 assigned the evaluation word 1100 "undesirable" does not satisfy the factor 602 belonging to the "C" group.

[0053] 14 is an explanatory diagram showing an example of generation of a new option by the inflection suggestion generation unit 306. The inflection suggestion generation unit 306 provides the text generation AI 104 with a conditional statement 1300 for generating a new option as a query, and obtains a new option 1400 as an answer statement.

[0054] 15 is an explanatory diagram showing an example of displaying a new option 1400 on the personal device 103. The consensus building support apparatus 101 transmits the new option 1400 to the personal device 103 and the existing option group 311. The personal device 103 displays a display screen 1500. The display screen 1500 displays the new option 1400 and the existing option group 311.

[0055] 16 is an explanatory diagram showing another example 1 of score calculation. In Fig. 16, the factor classification unit 304 performs principal component analysis of variables (three variables in this example because there are three participants) on participant-specific evaluation results 1601, and calculates a first principal component score as score 1602.

[0056] 17 is an explanatory diagram showing another example of score calculation 2. In Fig. 17, the factor classification unit 304 performs principal component analysis of variables (50 variables in this example because there are 50 participants) on participant-specific evaluation results 1701, and calculates a first principal component score as score 1702.

[0057] In reality, it may be necessary to reach a consensus with participants online regarding factors 602 or questions with many participants (evaluation results 711). Furthermore, when the questions become more complex, it is difficult to manually generate highly valid scores. Therefore, it is effective to use principal component analysis to calculate the axis that has the greatest influence on the participants' factors 602, and then calculate the first principal component score as the scores 1600 and 1702.

[0058] 18 is a two-dimensional scatter plot with the first and second principal components in the principal component analysis shown in FIG. 17 as axes. In the two-dimensional scatter plot 1800, the first principal component has a larger variance than the second principal component, and therefore serves as the axis that most clearly shows the participant's evaluation 710 of the factor group 312. Therefore, the factor classification unit 304 regards the axis of the first principal component as the axis that has the greatest influence on the participant's factor 602, and calculates the first principal component score as the score 1702.

[0059] 19 is an explanatory diagram showing another example 1 of group classification by the factor classification unit 304. The score 1901 is calculated using the same calculation method as the score 1001. Since the participant-specific evaluation result 1601 is different from the participant-specific evaluation result 900 in FIG. 10 , the score 1901 will have a different value from the score 1001.

[0060] The factor classification unit 304 performs cluster analysis such as the k-means method based on the score 1901, and classifies the factor group 312 into a specified number of groups. In Fig. 19, the factor classification unit 304 classifies the factors 312 into three groups, groups A to C, in descending order of the center of gravity (1.8, -0.63, -3) of each cluster in the k-means method (classification 1902).

[0061] 20 is an explanatory diagram showing another example 2 of group classification by the factor classification unit 304. When the number of factors 602 and participants (evaluation results 711) increases, it is possible that stakeholders may reach a consensus with participants online about an issue with a large number of factors 602 and participants (evaluation results 711). Furthermore, when issues become so complex, it is difficult to manually assign a highly valid score. Therefore, a method of clustering factors 602 with similar evaluations 710 through cluster analysis is effective.

[0062] The factor classification unit 304 performs principal component analysis using variables that are quantified from the three-level evaluation 710 of the participant-specific evaluation results 1701 (◯ = "1", △ = "0", × = "-1"), and calculates the first principal component score as score 1702.

[0063] The factor classification unit 304 performs cluster analysis such as the k-means method based on the score 1702, and classifies the factor group 312 into a specified number of groups. In Fig. 20, the factor classification unit 304 classifies the factors 312 into five groups, A to E, in descending order of the center of gravity of each cluster in the k-means method (3.03, 1.6, 0.23, -0.94, -2.13) (classification 2000).

[0064] <Example of Generating Factor Group 312> Fig. 21 is an explanatory diagram showing an example of generating the factor group 312 in the database 102. The condition generating unit 305 generates a condition statement 2101 for listing the advantages and disadvantages of one of the existing option groups 311 in the database 102 (in Fig. 21, "Attracting Businesses" is used as an example).

[0065] The synthesis proposal generation unit 306 provides the text generation AI 104 with a conditional statement 2100 for listing the advantages and disadvantages of attracting businesses as a query, and obtains the advantages and disadvantages that are factors in attracting businesses, which are the existing options, as an answer statement 2102. The answer statement 2102 is sent to the personal device 103 and displayed on the personal device 103.

[0066] In FIG. 21, the conditional statement 2101 is the advantages and disadvantages of the option ("attracting businesses"), but it may be at least one of the advantages and disadvantages.

[0067] <Another Example of Generating New Options> Fig. 22 is an explanatory diagram showing another example of generating new options. The condition generating unit 305 selects multiple options (in Fig. 22, as an example, "attracting businesses" and "supporting entrepreneurship") from the existing option group 311 in the database 102, and generates a condition statement 2201 that satisfies all of the multiple options (attracting businesses and supporting businesses).

[0068] The synopsis proposal generation unit 306 provides the text generation AI 104 with a conditional statement 2201 that satisfies all of the multiple options (attracting businesses and supporting businesses) as a query, and obtains a new option that satisfies the multiple options as an answer statement 2202. The answer statement 2202 is sent to the personal device 103 and displayed on the personal device 103.

[0069] In the example of FIG. 22 , three options are generated as new options from multiple options: "support programs and loan systems for entrepreneurs," "marketing support for local industries," and "tax incentives." The consensus building support device 101 adds the newly generated options, "support programs and loan systems for entrepreneurs," "marketing support for local industries," and "tax incentives," to the existing option group 311 in the database 102. Furthermore, the consensus building support device 101 sets the factors 602 of "business attraction" and "business support" to the factors 602 of each of the newly generated options, "support programs and loan systems for entrepreneurs," "marketing support for local industries," and "tax incentives."

[0070] Then, the consensus building support device 101 deletes the entries of “attracting businesses” and “supporting businesses” and their factors 602 from the database 102 .

[0071] In this case, the factors 602 of the new options become the factors 602 of the deleted options (business attraction and business support). Then, for each of the new options t, the consensus building support device 101 retains the factors 602 to which the evaluation word 1100 "desirable" is assigned, as shown in FIG. 12, and deletes the factors 602 to which the evaluation word 1100 "undesirable" is assigned. In this way, the consensus building support device 101 can automatically set the factors 602 for the new options.

[0072] In addition, when generating the existing group of options 311 and their factors 602, the consensus building support device 101 can also organize them in advance using the text generation AI 104.

[0073] 23 is an explanatory diagram showing an example of dividing a policy. For example, the condition generating unit 305 selects an option (in FIG. 23, for example, "attracting businesses") from the existing option group 311 in the database 102, and generates a condition statement 2301 for dividing the option (attracting businesses) into two policies.

[0074] The synthesis proposal generation unit 306 provides the text generation AI 104 with a conditional statement 2301 for dividing the option (attracting businesses) into two measures as a query, and acquires the two measures as an answer statement 2302. The consensus building support device 101 can refer to the answer statement 2302 through a user operation and make modifications to the existing option group 311 in the database 102, such as adding "tax incentives" or "infrastructure development" or deleting "attracting businesses."

[0075] In addition, when generating the existing group of options 311 and their factors 602, the consensus building support device 101 can also organize them in advance using the text generation AI 104.

[0076] 24 is an explanatory diagram showing an example of dividing the factor 602. For example, the condition generating unit 305 selects a factor (in FIG. 24, the factor for attracting businesses, "increasing local employment") from the group of factors 312 in the database 102, and generates a condition statement 2401 for dividing the selected factor into two factors.

[0077] The synthesis proposal generation unit 306 provides the text generation AI 104 with a conditional statement 2401 for dividing a factor ("increasing local employment") into two factors as a query, and acquires the two factors as a response statement 2402. The consensus building support device 101 can refer to the response statement 2402 through a user operation and make modifications to the group of factors 312 in the database 102, such as adding "job creation through corporate expansion" or "expanding local employment opportunities through an increase in employees" or deleting "increasing local employment."

[0078] 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.

[0079] 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.

[0080] 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).

[0081] 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 consensus building support device that can access a database that stores a group of options to be taken for a policy and a group of factors for each option in the group of options to be established, the consensus building support device having a processor that executes a program and a storage device that stores the program, wherein the processor executes the following: an acquisition process that acquires, for each participant, evaluation results obtained by evaluating each factor in the group of factors on a multiple-level scale; a classification process that classifies the group of factors into a plurality of groups including a first group to which a first factor that is positively evaluated belongs and a second group to which a second factor that is negatively evaluated belongs, based on the evaluation results for each participant acquired by the acquisition process; and an output process that outputs the classification results obtained by the classification process.

2. A consensus building support device as described in claim 1, wherein the database stores a group of evaluation words having at least one of positive evaluation words and negative evaluation words for evaluating the factors, and in the classification process, if the group of evaluation words contains the positive evaluation word, the processor assigns the positive evaluation word to the first group, and if the group of evaluation words contains the negative evaluation word, the processor assigns the negative evaluation word to the second group.

3. A consensus building support device as described in claim 1, characterized in that, in the classification process, the processor calculates a score for each factor by quantifying the evaluation results, and classifies the group of factors into multiple groups based on the scores.

4. A consensus building support device as described in claim 3, characterized in that, in the classification process, the processor quantifies the evaluation results and performs principal component analysis to calculate a first principal component score for each factor as the score, and classifies the group of factors into the multiple groups based on the score.

5. A consensus building support device as described in claim 3, characterized in that, in the classification process, the processor classifies the group of factors into the multiple groups by performing cluster analysis based on the scores.

6. A consensus building support device as described in claim 4, characterized in that in the classification process, the processor classifies the group of factors into the multiple groups by performing cluster analysis based on the scores.

7. A consensus building support device as described in claim 1, characterized in that the processor executes a generation process to generate a conditional statement inquiring about a new measure different from the measure based on the classification result by the classification process, and in the output process, the processor outputs the conditional statement generated by the generation process.

8. A consensus building support device as described in claim 7, characterized in that, in the generation process, the processor generates a conditional statement that queries about a new measure that satisfies the first factor and does not satisfy the second factor.

9. A consensus building support device as described in claim 8, wherein the processor executes an answer sentence acquisition process to acquire an answer sentence indicating the new measure by inputting a conditional statement into a statement generation function which, when given an input statement, interprets the input statement and outputs an output statement for the input statement, and in the output process, the processor outputs the answer statement acquired by the answer sentence acquisition process.

10. A consensus building support device as described in claim 8, characterized in that in the generation process, the processor generates a conditional statement that queries at least one of the advantages and disadvantages of the options.

11. A consensus building support device as described in claim 10, wherein the processor executes an acquisition process to acquire an answer sentence indicating at least one of the advantages and disadvantages of the option by inputting the conditional statement into a statement generation function which, when given an input statement, interprets the input statement and outputs an output statement for the input statement, and in the output process, the processor outputs the answer statement acquired by the acquisition process.

12. A consensus building support device as described in claim 7, characterized in that, in the generation process, the processor generates a conditional statement that queries for a new option that satisfies multiple options in the group of options.

13. A consensus building support device as described in claim 12, wherein the processor executes an acquisition process to acquire an answer sentence indicating the new option by inputting the conditional statement into a statement generation function which, when given an input statement, interprets the input statement and outputs an output statement for the input statement, and in the output process, the processor outputs the answer sentence acquired by the acquisition process.

14. A consensus building support method executed by a consensus building support device that can access a database that stores a group of options to be taken for a policy and a group of factors for each option in the group of options to be established, the consensus building support method comprising: a processor that executes a program; and a storage device that stores the program, wherein the processor executes: an acquisition process that acquires, for each participant, an evaluation result obtained by evaluating each factor in the group of factors on a multiple-level scale; a classification process that classifies the group of factors into a plurality of groups including a first group to which a first factor that is positively evaluated belongs and a second group to which a second factor that is negatively evaluated belongs, based on the evaluation result for each participant acquired by the acquisition process; and an output process that outputs the classification result obtained by the classification process.

15. A consensus building support program characterized by having a processor that can access a database that stores a group of options to be taken for a policy and a group of factors for each option in the group of options to be established execute the following steps: an acquisition process that acquires, for each participant, evaluation results obtained by evaluating each factor in the group of factors on a multiple-level scale; a classification process that classifies the group of factors into a plurality of groups including a first group to which a first factor that is evaluated positively belongs and a second group to which a second factor that is evaluated negatively belongs, based on the evaluation results for each participant acquired by the acquisition process; and an output process that outputs the classification results obtained by the classification process.