Evaluation support system, evaluation support method, and evaluation support program

Abstract

To provide an evaluation support system, an evaluation support method and an evaluation support program for generating a logic model in a measure and evaluating an impact.SOLUTION: The support device 20 includes a control unit 21 connected to the user device 10. The control unit 21 includes a logic model generation unit 212, an index generation unit 213, and an evaluation unit 214. The logic model generation unit 212 generates elements constituting each hierarchy of the logic model from the evaluation target designation information acquired from the user device 10, and estimates a causal relationship between the elements. The index generation unit 213 generates an index for evaluating the element from the evaluation target designation information. The evaluation unit 214 predicts an index value of an index of each layer of the logic model and outputs the index value to the user device 10.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an evaluation support system, an evaluation support method, and an evaluation support program for evaluating the impact of a business on society. [Background technology]

[0002] Impact evaluations are conducted to measure the changes (impact) that a project has brought to a target society. Project evaluation devices that can effortlessly evaluate the social impact of a project are also under consideration (for example, Patent Document 1). The project evaluation device described in this document generates evaluation results of the social impact of a project based on the results and outcomes of the project.

[0003] A logic model is used when performing impact assessment. This logic model clearly shows the logical causal relationships until the measures of a business, project, etc. achieve their objectives. Support devices for assisting in the creation of such logic models have also been considered (for example, Patent Document 2). The logic model creation support device described in this document stores a logic model, which is a network structure composed of nodes having indicators and edges indicating the connection between the indicators indicated by the two nodes. A first logic model is generated based on the input first node and first edge. Furthermore, when a first node in the first logic model is specified, a similar indicator similar to the first indicator of the first node is identified from the logic model and the similar indicator is output. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-77772 [Patent Document 2] Patent No. 7412585 Summary of the Invention [Problem to be solved by the invention]

[0005] However, when generating a logic model, it is difficult to create indicators for evaluating the impact. [Means for solving the problem]

[0006] An evaluation support system for solving the above-mentioned problems includes a control unit connected to a user device, the control unit including: a logic model generation unit that generates elements constituting each layer of a logic model from evaluation target specification information acquired from the user device and estimates causal relationships among the elements, an index generation unit that generates indexes for evaluating the elements from the evaluation target specification information, and an evaluation unit that predicts index values ​​of the indexes for each layer of the logic model and outputs the predicted index values ​​to the user device. [Effects of the Invention]

[0007] According to the present disclosure, it is possible to generate a logic model for a policy and evaluate its impact. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is an explanatory diagram of an evaluation support system according to a first embodiment. [Figure 2] FIG. 2 is an explanatory diagram of a hardware configuration of the first embodiment. [Figure 3] FIG. 2 is an explanatory diagram of a logic model according to the first embodiment. [Figure 4] FIG. 3 is an explanatory diagram of a processing procedure of the evaluation support processing according to the first embodiment. [Figure 5] FIG. 10 is an explanatory diagram of a processing procedure of the evaluation support processing according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] (First embodiment) A first embodiment of an evaluation support system, an evaluation support method, and an evaluation support program will be described with reference to Figures 1 to 4. In this embodiment, it is assumed that a logic model is generated for evaluating the impact of measures such as business, activities, and projects in an organization.

[0010] As shown in FIG. 1, the evaluation support system of this embodiment uses a user device 10, a database system 15, a support device 20, and an AI system 30, all of which are connected to one another via a network.

[0011] (Example of hardware configuration) FIG. 2 shows an example of the hardware configuration of an information processing device H10 that functions as the user device 10, the database system 15, the support device 20, the AI ​​system 30, and the like.

[0012] The information processing device H10 includes a communication device H11, an input device H12, a display device H13, a storage device H14, and a processor H15. Note that this hardware configuration is an example, and the information processing device H10 may include other hardware.

[0013] The communication device H11 is an interface that establishes a communication path with another device and executes data transmission and reception, and is, for example, a network interface or a wireless interface.

[0014] The input device H12 is a device that accepts input from a user, etc., and is, for example, a mouse, a keyboard, etc. The display device H13 is a display, a touch panel, etc. that displays various information.

[0015] The storage device H14 is a storage device that stores data and various programs for executing various functions of the user device 10, the database system 15, the assistance device 20, and the AI ​​system 30. Examples of the storage device H14 include a ROM, a RAM, a hard disk, etc.

[0016] The processor H15 controls each process in the user device 10, the database system 15, the assistance device 20, and the AI ​​system 30 using programs and data stored in the storage device H14. Examples of the processor H15 include a CPU and an MPU. The processor H15 loads programs stored in a ROM or the like into a RAM and executes various processes corresponding to the various processes. For example, when an application program of the user device 10, the database system 15, the assistance device 20, or the AI ​​system 30 is launched, the processor H15 runs a process that executes each process described below.

[0017] The processor H15 is not limited to a processor that performs all of its processing using software. For example, the processor H15 may include a dedicated hardware circuit (e.g., an application-specific integrated circuit (ASIC)) that performs hardware processing for at least part of the processing it performs. That is, the processor H15 may be configured, for example, with the following:

[0018] [1] One or more processors that operate according to a computer program (software). [2] One or more dedicated hardware circuits that perform at least some of the various processes [3] Circuits containing combinations of these The processor includes a CPU and memory, such as RAM and ROM, that stores program code or instructions configured to cause the CPU to perform processes. Memory, or computer-readable media, includes any available media that can be accessed by a general-purpose or special-purpose computer.

[0019] (Logic Model) First, the generation of a logic model will be explained using FIG. When generating a logic model, the evaluation object D1 for which impact evaluation will be performed is identified. This evaluation object D1 can be, for example, an organization's business, activities, or measures such as projects. When evaluating a project, the project's purpose, goals, expectations, etc. are identified. The project purpose is what the project wants to achieve as a whole. The project goal is the specific indicators set to achieve the goal. The project expectations are what the project stakeholders expect from the project. To identify this evaluation object D1, for example, a business report can be used.

[0020] Next, the layers that make up logic model D2 are created from evaluation target D1. This logic model D2 is composed of the layers of "input D21 (input)," "activity D22 (activities)," "output D23 (output)," "outcome D24 (results)," and "impact D25 (influence)." Here, "input D21" is defined as the resources required to implement the project. "activity D22" is defined as the activity to achieve results by utilizing input D21. "output D23" is defined as the results directly obtained through activity D22. "outcome D24" is defined as the medium- to long-term results brought about by output D23. "impact D25" is defined as the final social and environmental impact brought about by outcome D24.

[0021] Then, identify the elements D3 to be used at each level of the logic model. The "Input D21" element D3 includes, for example, the resources required to implement the project (human resources, funds, equipment, information, etc.). The "Activity D22" element D3 includes, for example, activities to achieve results using Input D21 (training, events, consulting, product development, etc.). The "Output D23" element D3 includes, for example, the direct results obtained through the activity (reports, materials, software, products, etc.). The "Outcome D24" element D3 includes, for example, the medium- to long-term results brought about by Output D23 (improvement of knowledge and skills, behavioral change, institutional reform, etc.). The "Impact D25" element D3 includes, for example, the final social and environmental impacts brought about by Outcome D24 (poverty reduction, environmental improvement, economic growth, etc.).

[0022] Then, indicators D4 are established for each element. These indicators D4 are used to evaluate each element of the logic model D2 (input D21 to impact D25). "Input D21" indicators quantify the quantity and quality of resources invested (human resources, funds, facilities, information, etc.). "Activity D22" indicators quantify the content and scale of the activities implemented (number of events, number of participants, duration of presentation, content provided, number of materials created, number of copies distributed, etc.). "Output D23" indicators quantify the direct results of the activities (number of reports, number of materials, software version, functions, number of products, number of sales, etc.). "Outcome D24" indicators quantify the medium- to long-term results of output D23 (degree of knowledge and skill acquisition, level of understanding, degree of behavioral change, progress of institutional reform, etc.). The "Impact D25" indicators include those that quantify the ultimate social and environmental impact of a project (environmental pollution indicators, biodiversity indicators, economic growth rate, number of jobs created, etc.). Then, the index values ​​D5 of the respective indices are set by quantifying the respective indices D4.

[0023] (Functions of each information processing device) Using FIG. 1, the functions of a user device 10, a database system 15, a support device 20, and an AI system 30 used in this evaluation support system will be described.

[0024] The user device 10 is a computer terminal used by a user of this system. The database system 15 is a computer system that can aggregate, store, manage, and refer to various information. This database system 15 stores information related to measures such as business, activities, projects, etc., such as corporate financial information, product information, industry and product trend information, media information, and IR materials.

[0025] The support device 20 is a computer system for utilizing the generation AI. The support device 20 includes a control unit 21, a performance information storage unit 22, and an evaluation target information storage unit .

[0026] The control unit 21 performs the processes described below (processes including a management stage, a logic model generation stage, an index generation stage, an evaluation stage, etc.) By executing an evaluation support program for this purpose, the control unit 21 functions as a management unit 211, a logic model generation unit 212, an index generation unit 213, an evaluation unit 214, etc.

[0027] The management unit 211 executes a process of acquiring various information from the user device 10. The logic model generation unit 212 executes a process for generating a logic model using the AI ​​system 30. The logic model generation unit 212 creates a logic model generation prompt for generating a logic model, inputs it to the AI ​​system 30, and obtains a response.

[0028] The index generation unit 213 executes a process of generating an index using the AI ​​system 30. The index generation unit 213 creates an index generation prompt for generating an index, inputs it to the AI ​​system 30, and obtains a response. The evaluation unit 214 executes a process for evaluating the validity of the impact of the created logic model.

[0029] The performance information storage unit 22 stores performance management information related to logic models created in the past. The performance management information is registered when a logic model is acquired or created. This performance management information stores evaluation target information and logic model information.

[0030] The evaluation target information is information for identifying the target of an existing logic model created in the past. The logic model information includes information about the elements that make up each layer of an existing logic model created in the past, the indicators for each element, and the indicator values ​​for each indicator.

[0031] Evaluation target information about the evaluation target is recorded in the evaluation target information storage unit 23. This evaluation target information is registered when it is acquired from the user device 10. The evaluation target information records evaluation target designation information (text information), context information, and logic model candidate information about the evaluation target.

[0032] The evaluation target designation information is information about the evaluation target acquired from the user device 10. The context information is additional information that is input to the AI ​​system 30 to allow it to understand the evaluation target specification information. The logic model candidate information includes elements and indicators of each hierarchy created using the AI ​​system 30, and the index values ​​of each indicator.

[0033] The AI ​​system 30 is a computer system that utilizes generative AI technology (large-scale language model) to generate responses to prompts. The AI ​​system 30 is configured according to computational models and parameters, which are elements that determine the structure of the generative AI.

[0034] (Evaluation support processing) The evaluation support process will be described with reference to FIG. First, the control unit 21 of the support device 20 executes a text acquisition process (step S11). Specifically, the management unit 211 of the control unit 21 acquires evaluation target designation information related to the company or project for which a logic model is to be created. For example, a business report or the like can be used as the evaluation target designation information. Then, the management unit 211 records the acquired evaluation target designation information (text information) in the evaluation target information storage unit 23.

[0035] Next, the control unit 21 of the assistance device 20 executes a context acquisition process (step S12). Specifically, the management unit 211 of the control unit 21 acquires context information to be input to the generation AI from the database system 15. This context information is used in search expansion generation. In this search expansion generation, a pre-trained language model and an external knowledge base are combined. Here, in the prompt, words included in the evaluation target designation information are used to search for related information in the knowledge base. Based on the search results, a response is acquired using the language model. Here, context information related to keywords included in the acquired evaluation target designation information is acquired from the database system 15 (external knowledge base). For example, the context information can be company financial information, product information, industry and product trend information, media information, etc.

[0036] Next, the control unit 21 of the assistance device 20 executes a process for generating elements usable in the logic model (step S13). Specifically, the logic model generation unit 212 of the control unit 21 generates a logic model generation prompt for generating a logic model. This logic model generation prompt includes a definition body and a format of the logic model. The definition body is configured by definitions of each hierarchical level (input to impact) of the logic model. The format is configured by a format for describing the logic model (input → activity → output → outcome → impact). This logic model generation prompt includes an instruction to generate elements to be included in the logic model using evaluation target specification information and context information. Note that each element is usable at any hierarchical level. Then, the logic model generation unit 212 inputs the logic model generation prompt to the AI ​​system 30 using search expansion and generation (RAG). Then, the logic model generation unit 212 obtains a response from the AI ​​system 30, including elements usable at each hierarchical level of the logic model. In this case, each element is connected by a causal relationship and distributed to each hierarchical level according to the definition body and format.

[0037] Next, the control unit 21 of the support device 20 executes an index extraction process for each element (step S14). Specifically, the index generation unit 213 of the control unit 21 identifies an index that can be used for each acquired element. Here, the index generation unit 213 identifies an index for an element from evaluation target information and past cases in the performance information storage unit 22.

[0038] Next, the control unit 21 of the support device 20 executes an index value generation process (step S15). Specifically, the evaluation unit 214 of the control unit 21 generates an index value generation prompt that predicts index values ​​explicitly or implicitly described in the evaluation target designation information or context information. If the evaluation target designation information or context information includes index values, the index value generation prompt includes an instruction to acquire these index values. On the other hand, if the evaluation target designation information or context information does not include an index value corresponding to the indicator, the index value generation prompt includes an instruction to calculate a measure for calculating the index value using a "maturity model." Here, a maturity model, for example, is selected as an evaluation model for evaluating a program or project. For example, CMMI (Capability Maturity Model Integration) can be used as the maturity model. In CMMI, maturity levels are expressed as an initial "Level 1," a managed "Level 2," a defined "Level 3," a quantitatively managed "Level 4," and an optimized "Level 5." The evaluation unit 214 then inputs the index value generation prompt to the AI ​​system 30 and acquires an index value for each element as a response. When using a maturity model, the evaluation unit 214 acquires, from the AI ​​system 30, a level that is one step higher than the current state as an index value.

[0039] Next, the control unit 21 of the support device 20 executes a process for evaluating the validity of the index value for each element (step S16). Specifically, the evaluation unit 214 of the control unit 21 associates the logic model with the evaluation target information and records it in the evaluation target information storage unit 23. Then, the evaluation unit 214 outputs the index value for each element, including the impact, to the display device H13 of the user device 10. Here, if the index value for each element is not valid, the management unit 211 receives a re-execution instruction from the user device 10. In this case, the evaluation unit 214 instructs the logic model generation unit 212 to recreate the logic model. In this case, the logic model generation unit 212 receives a new logic model, indexes, index values, etc. from the AI ​​system 30. On the other hand, if the management unit 211 receives an end instruction from the user device 10, the evaluation unit 214 registers the evaluation target information (evaluation target designation information, context information) and the logic model in the performance information storage unit 22.

[0040] (Operation of the embodiment) Since a logic model is generated in accordance with the evaluation target designation information, impact evaluation can be performed using the logic model.

[0041] (Effects of the first embodiment) (1-1) In this embodiment, the control unit 21 of the assistance device 20 executes a text acquisition process (step S11) and a context acquisition process (step S12). This enables the AI ​​system 30 to perform search expansion and generation.

[0042] (1-2) In this embodiment, the control unit 21 of the support device 20 executes a process for generating elements that can be used in a logic model (step S13), thereby making it possible to identify elements to be used in the logic model.

[0043] (1-3) In this embodiment, the control unit 21 of the support device 20 executes an index extraction process (step S14) and an index value generation process (step S15) for each element, thereby identifying the indexes to be used in the logic model.

[0044] (1-4) In this embodiment, the control unit 21 of the support device 20 executes a process for evaluating the validity of the index value for each element (step S16), thereby determining the validity of the logic model.

[0045] (Second embodiment) Next, a second embodiment of an evaluation support system, an evaluation support method, and an evaluation support program will be described with reference to Fig. 5. In the first embodiment, a case where a logic model is generated using evaluation target designation information and context information was described. In the second embodiment, the configuration is modified so that a logic model is generated by including information that is missing from the evaluation target designation information and context information. The same reference numerals are used to designate the same parts as in the first embodiment, and detailed descriptions thereof will be omitted.

[0046] First, the control unit 21 of the assistance device 20 executes the text acquisition process (step S21) and the context acquisition process (step S22) in the same manner as steps S11 and S12. Next, the control unit 21 of the support device 20 executes a process of identifying logic model candidates (step S23). Specifically, the logic model generation unit 212 of the control unit 21 uses the evaluation target designation information as a query to search the performance information storage unit 22. In this case, the logic model generation unit 212 converts the information included in the evaluation target designation information into a vector expression and acquires a predetermined number of logic model candidates with high similarity.

[0047] Then, the control unit 21 of the support device 20 executes hearing item information acquisition processing for each logic model candidate (step S24). Specifically, the logic model generation unit 212 of the control unit 21 inputs a hearing item prompt including evaluation target designation information, context information, and logic model candidate to the AI ​​system 30. This hearing item prompt includes an instruction to list hearing items for extracting information that is missing from the evaluation target designation information and context information in order to obtain the logic model candidate. Then, the logic model generation unit 212 acquires the hearing items as a response from the AI ​​system 30, associates them with the logic model candidate, and records them in the evaluation target information storage unit 23. Then, the processing is repeated until it is completed for all logic model candidates.

[0048] Next, the control unit 21 of the support device 20 executes a logic model candidate selection process (step S25). Specifically, the logic model generation unit 212 of the control unit 21 selects a logic model candidate. Here, for example, a logic model candidate with the fewest hearing items is selected. Note that, among the logic model candidates, multiple logic model candidates may be selected in order of the fewest hearing items.

[0049] Next, the control unit 21 of the assistance device 20 executes a process of outputting hearing items for each logic model candidate (step S26). Specifically, the logic model generation unit 212 of the control unit 21 outputs the hearing items of the logic model candidate to the user device 10. Here, the logic model generation unit 212 may input a prompt to the AI ​​system 30 to create a question or message for acquiring the hearing items.

[0050] Next, the control unit 21 of the support device 20 executes a process of identifying elements that can be used in the logic model (step S27). Specifically, the logic model generation unit 212 of the control unit 21 identifies elements of logic model candidates.

[0051] Next, the control unit 21 of the support device 20 executes, similarly to steps S14 to S16, an index extraction process for each element (step S28), an index value generation process (step S29), and an index value validity evaluation process for each element (step S30).

[0052] (Effects of the second embodiment) According to the second embodiment, in addition to the effects (1-1), (1-3), and (1-4), the following effects can be obtained.

[0053] (2-1) In this embodiment, the control unit 21 of the support device 20 executes a process of identifying logic model candidates (step S23), thereby enabling a proven logic model to be identified as a candidate.

[0054] (2-2) In this embodiment, the control unit 21 of the support device 20 executes an information acquisition process for hearing items for each logic model candidate (step S24). This allows the control unit 21 to acquire missing information and generate a logic model.

[0055] (2-3) In this embodiment, the control unit 21 of the support device 20 executes a process of selecting logic model candidates (step S25), thereby enabling the inclusion of logic model candidates.

[0056] (2-4) In this embodiment, the control unit 21 of the support device 20 executes a process of outputting hearing items for each logic model candidate (step S26), thereby making it possible to grasp the contents of the hearing to be carried out.

[0057] This embodiment can be modified as follows: This embodiment and the following modifications can be combined and implemented within the scope of technical compatibility. In the above embodiments, a user device 10, a database system 15, a support device 20, and an AI system 30 are used. However, the hardware configuration is not limited to these. For example, the database system 15 and the AI ​​system 30 may be provided within the support device 20. In each of the above embodiments, the control unit 21 of the support device 20 executes a text acquisition process (step S11). Here, evaluation target designation information relating to the company or project for which a logic model is to be created is acquired. In this case, the name of the company or project to be evaluated may be acquired as the evaluation target designation information, and information (HTML files, PDF files, etc.) may be acquired by searching (crawling) the Internet, etc. In this case, the amount of information (number of files) to be acquired may be specified in advance.

[0058] In each of the above embodiments, the logic model D2 is composed of layers from "input D21" to "impact D25." The layers are not limited to these. For example, a "task" may be included. Here, a task is defined as a problem that arises in achieving a result, a hurdle that must be overcome, and a countermeasure for the problem. Alternatively, the user may be allowed to select any definition body of each hierarchy that constitutes the logic model D2. In this case, a plurality of definition bodies are prepared in the support device 20. Then, the control unit 21 generates a prompt using the definition body of each hierarchy selected by the user device 10.

[0059] In each of the above embodiments, the control unit 21 of the support device 20 executes the process of generating index values ​​(steps S15 and S29). Here, index values ​​are acquired from evaluation target specification information or context information, or are predicted using a maturity model as an evaluation model. Alternatively, index values ​​of elements related to a specific element may be predicted from the index values ​​of the specific element. For example, index values ​​of elements for which explicit or implicit index values ​​could not be identified from the index values ​​of elements with causal relationships are estimated and interpolated. Furthermore, the evaluation model is not limited to the maturity model (CMMI). For example, an original evaluation scale may be defined.

[0060] In each of the above embodiments, the control unit 21 of the support device 20 executes a process for evaluating the validity of the index value for each element (steps S16 and S30). Specifically, the evaluation unit 214 of the control unit 21 outputs the index value for each element to the display device H13 of the user device 10. Alternatively, the evaluation unit 214 may evaluate the validity of the index value for each element. In this case, the validity of each index value is determined based on the causal relationship between the elements. For example, a valid range of the index value of a subsequent element is estimated relative to the index value of a preceding element, and the index value of the subsequent element is determined to be valid if it is within the valid range. Furthermore, the valid range of the index value of each element may be estimated depending on the scale of the evaluation target. In addition, the index values ​​of other companies in the same industry may be output as supplementary information. In this case, the control unit 21 of the support device 20 obtains index values ​​of each element for other companies in the same industry specified by the user, for example, through an Internet search, and outputs the results of comparing them with the evaluation target. In addition, other companies in the same industry may be identified using industry information for the company or project for which the logic model is being created.

[0061] In each of the above embodiments, logic model generation prompts and index value generation prompts are used in search expansion generation. Alternatively, machine learning may be performed to generate a predictive model. In this case, evaluation target specification information and a logic model are used as training information, and machine learning is performed to generate a learning model that outputs a logic model (elements, indexes, and index values ​​for each layer) or fine-tuning of a large-scale language model. This allows impact evaluation to be performed without using individual prompts.

[0062] In the second embodiment, the control unit 21 of the support device 20 executes a process of identifying logic model candidates (step S23). Specifically, the logic model generation unit 212 of the control unit 21 searches the performance information storage unit 22 using the evaluation target designation information as a query. Alternatively, logic model candidates may be generated using a logic model generation prompt. In this case, it is acceptable for elements to be missing in some hierarchical levels. Then, the logic model generation unit 212 identifies the missing elements as hearing items in the hearing item prompt.

[0063] In the second embodiment, the control unit 21 of the support device 20 executes a logic model candidate selection process (step S25). Here, a logic model candidate with a small number of hearing items is selected. Alternatively, a logic model candidate selected by the user device 10 may be acquired. Alternatively, a logic model candidate with a high feasibility may be selected. [Explanation of symbols]

[0064] 10...user device, 15...database system, 20...support device, 21...control unit, 211...management unit, 212...logic model generation unit, 213...index generation unit, 214...evaluation unit, 22...performance information storage unit, 23...evaluation target information storage unit, 30...AI system.

Claims

1. An evaluation support system that supports impact evaluation, comprising a control unit connected to a user device, The control unit a logic model generation unit that generates elements constituting each layer of a logic model from the evaluation target designation information acquired from the user device and estimates causal relationships of the elements; an index generating unit that generates an index for evaluating the element from the evaluation target designation information; an evaluation unit that predicts index values ​​of indexes of each layer of the logic model and outputs the predicted index values ​​to the user device.

2. The logic model generation unit Using a word included in the evaluation target designation information, a context related to the word is obtained; 2. The evaluation support system according to claim 1, wherein the logic model is generated using the context.

3. The evaluation support system described in claim 1 or 2, characterized in that the evaluation unit obtains index values ​​of indicators of elements at each level of the generated logic model from a database that stores existing logic models created in the past.

4. 4. The evaluation support system according to claim 3, wherein the evaluation unit predicts the index value using a maturity model when the index value cannot be identified in the database.

5. The evaluation support system described in claim 1, characterized in that the logic model generation unit, the index generation unit, and the evaluation unit are generated by performing machine learning using the evaluation target designation information, the elements that make up each layer of the logic model, and the index values ​​of the indexes as teaching information.

6. An evaluation support method for supporting impact evaluation using an evaluation support system including a control unit connected to a user device, comprising: The control unit generating elements constituting each layer of a logic model from the evaluation target designation information acquired from the user device, and estimating causal relationships among the elements; generating an index for evaluating the element from the evaluation target designation information; An evaluation support method comprising predicting index values ​​of indexes of each layer of the logic model and outputting the predicted index values ​​to the user device.

7. An evaluation support program that supports impact evaluation using an evaluation support system including a control unit connected to a user device, The control unit a logic model generation unit that generates elements constituting each layer of a logic model from the evaluation target designation information acquired from the user device and estimates causal relationships between the elements; an index generating unit that generates an index for evaluating the element from the evaluation target designation information; and an evaluation support program for functioning as an evaluation unit that predicts index values ​​of indexes of each layer of the logic model and outputs the values ​​to the user device;

Images