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

Abstract

This invention provides an information processing program, an information processing device, and an information processing method that can assist companies in the process of responding to questionnaires created in any format. [Solution] The information processing program related to the disclosed technology causes a computer to execute the following steps: an questionnaire acquisition procedure (ST110) for acquiring questionnaire data; a conversion procedure (ST120) for acquiring data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format; and an answer acquisition procedure (ST140) for acquiring data indicating answers to questions included in a modified questionnaire, using company information held by a predetermined company for answering questions included in the modified questionnaire and data indicating the modified questionnaire.

Description

Technical Field

[0001] The disclosed technology relates to an information processing program, an information processing apparatus, and an information processing method.

Background Art

[0002] Conventionally, technologies have been proposed for a company to obtain various information from a trading partner company using a computer. For example, in Patent Document 1, a business entity (company) in a supply chain can use a computer to obtain, from a plurality of business entities constituting an upstream process or a downstream process of the supply chain, information such as a target reduction rate of greenhouse gas or information related to greenhouse gas emissions.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Technologies exist that allow a company in a supply chain (hereinafter referred to as the "acquiring company") to use a computer to acquire various information from a company located in an upstream or downstream process in that supply chain (hereinafter referred to as the "acquiring company"), such as the technology disclosed in Patent Document 1. However, if, for example, the acquiring company uses a different computer system than the one used by the acquiring company, it may be difficult for the acquiring company to acquire information through the system. In such cases, the acquiring company may send data such as a prescribed form or questionnaire (hereinafter collectively referred to as the "questionnaire") to the acquiring company and request that the acquiring company input its answers to the questionnaire. Here, the types of questionnaires that the acquiring company sends to the acquiring company as data are diverse, and the format may also be arbitrary, in which case it becomes a burden for the acquiring company to answer these questionnaires.

[0005] The disclosed technology aims to solve the above-mentioned problems and provides an information processing program, information processing device, and information processing method that can support companies in the process of responding to questionnaires created in any format. [Means for solving the problem]

[0006] The information processing program relating to the disclosed technology causes a computer to execute the following: a questionnaire acquisition procedure for acquiring questionnaire data; a questionnaire conversion procedure for acquiring data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format, using the questionnaire data acquired in the questionnaire acquisition procedure; and a response acquisition procedure for acquiring data indicating the answers to the questions included in the modified questionnaire, using company information held by a predetermined company for answering questions included in the modified questionnaire identified by the data acquired in the questionnaire conversion procedure, and data indicating the modified questionnaire acquired in the questionnaire conversion procedure. [Effects of the Invention]

[0007] The information processing program relating to this disclosed technology has the above-described technical features and has the effect of supporting companies in the process of answering questionnaires created in any format. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram showing an example configuration of an information processing system related to the disclosed technology. [Figure 2] This figure shows the data structure for an information processing program related to the disclosed technology. [Figure 3] This figure shows an example of a questionnaire identified by the questionnaire data in the disclosed technology. [Figure 4] This figure shows an example of a modified questionnaire identified by the modified questionnaire data in the disclosed technology. [Figure 5] This figure shows an example of a modified questionnaire in which responses and reasons for those responses have been entered using the disclosed technology. [Figure 6] This diagram illustrates an example of the operation of the information processing device related to the disclosed technology and the flow of each piece of information. [Figure 7] This figure shows the hardware configuration of the information processing device related to the disclosed technology. [Figure 8] This is a flowchart showing the processing procedure of the information processing method related to the disclosed technology. [Modes for carrying out the invention]

[0009] While this disclosed technology is applicable to surveys in various business fields, this explanation will focus specifically on its application to ESG-related surveys, which have become increasingly important indicators for companies aiming for sustainable growth in recent years. For example, in recent years, ESG management has gained global attention as an essential management policy for companies to achieve sustainable growth, and an increasing number of companies are working to implement it. Furthermore, various disclosure systems are being developed in countries and regions around the world, requiring companies involved in activities within their respective areas to disclose information in accordance with disclosure standards from an ESG (Environment, Social, and Governance) perspective. There are also ESG rating agencies that evaluate whether organizations are conducting activities with ESG considerations in mind. In this way, ESG is gaining attention just like the SDGs. In addition to the term "ESG management" mentioned above, the term has become widely used in general, including compound words containing ESG such as "ESG investment," "ESG report," and "ESG analytics."

[0010] Several important terms used in the disclosed technology are defined below.

[0011] ESG Evaluation Criteria ESG evaluation items are defined as a single, identical item when they share common concepts or ideas, even if the wording of these evaluation items differs among different ESG rating agencies. ESG evaluation items include, for example, four items: ESG classification, code (number), major category, and minor category. In ESG evaluation criteria, the ESG classification indicates whether each data point falls under Environment, Social, or Governance. The letters used to represent the ESG classification are, for example, "E," "S," and "G." The codes (numbers) in ESG evaluation items are used to distinguish each subcategory, and each subcategory is assigned a different code. These codes that distinguish subcategories are also called "classification codes." The major and minor categories in ESG evaluation items are written names that make it easy for people to understand what kind of data each item represents that shows ESG information. These major and minor categories use generalized or standardized expressions of the ESG evaluation items used by each ESG rating agency. Examples of major categories include "General Environment," "Climate Change," "Water," "Resource Recycling," "Biodiversity," "Suppliers (Environment)," "Environmental Opportunities and Impacts," "Human Rights," "Labor Practices," "Diversity," "Human Resource Development," "Employee Safety and Health," "Corporate Citizenship," "Product Quality and Product Safety," "Corporate Governance," "Risk Management," and "Information Security and Privacy." Examples of minor categories include "Environmental Management Policies" under "General Environment," and "Policies to Address Climate Change," "Policies to Reduce Energy Consumption," and "Carbon Pricing" under "Climate Change."

[0012] ESG Evaluation Information In this specification, the term "ESG evaluation information" may include the following two types of information: One method is to provide each company with the evaluation results from various rating agencies, categorized by ESG evaluation item (hereinafter referred to as "company-specific evaluation information"). The evaluation results from each rating agency included in the company-specific evaluation information include, for example, whether the company meets or fails to meet the evaluation criteria adopted by each rating agency (hereinafter referred to as "meeting / failure"), and the ESG rating and score assigned by each rating agency (hereinafter simply referred to as "score"). Another piece of information is that which shows the evaluation criteria adopted by each rating agency in relation to each ESG evaluation item (hereinafter referred to as "evaluation criteria information"). ESG evaluation information includes, at a minimum, company-specific evaluation information. Furthermore, company-specific evaluation information and evaluation criteria information can be linked to corresponding ESG evaluation items and stored in the company information database described later for each company.

[0013] Evaluation Organization An evaluation institution is a third - party institution that evaluates a company's ESG efforts. Specifically, evaluation institutions include DJSI (Dow Jones Sustainability Index), FTSE, MSCI, Sustainalytics, CDP, etc. In addition, the evaluation methods used by evaluation institutions include the so - called questionnaire type, where the evaluation institution conducts interviews or questionnaires with the target company, and the so - called arbitrary rating type, where the evaluation institution conducts neither interviews nor questionnaires for evaluation. DJSI and CDP adopt the so - called questionnaire - type evaluation method. FTSE, MSCI, and Sustainalytics adopt the so - called arbitrary rating type evaluation method.

[0014] 《Disclosure Standards · Disclosure Standard Information》 Disclosure standards are standards that define the obligation to disclose ESG information by laws or frameworks. Examples of disclosure standards include GRI Standards, ISSB (Sustainability Disclosure Standards by the IFRS Foundation), etc. Disclosure standards are generally compiled as standard documents. In addition, the aforementioned evaluation criteria adopted by each evaluation institution are formulated based on such disclosure standards, and there is a close relationship between the two. Disclosure standard information is the electronic data of the standard document. The disclosure standard information of disclosure standards defined by various laws or frameworks is publicly available on the Internet.

[0015] Embodiment 1. Embodiment 1 shows the case where the disclosed technology is realized as the information processing device 100. FIG. 1 is a block diagram showing a configuration example of an information processing system according to the disclosed technology. As shown in FIG. 1, the information processing system is a system in which the information processing device 100 and the user terminal UT are connected via the Internet. Although Figure 1 shows three user terminals (UTs), the number of user terminals (UTs) can be any number of units, one or more. Also, although Figure 1 shows the information processing device 100 and the user terminals (UTs) connected via the Internet, this connection is not limited to the Internet; any telecommunications line may be used.

[0016] Information processing device 100 The information processing device 100 assists the user company (hereinafter referred to as the "user company") in preparing responses to questionnaires. The user company is, for example, a company located in an upstream or downstream process in a supply chain, and is a recipient company that responds to questionnaires sent as data from the acquiring company. The questionnaires that the user company responds to are, for example, questionnaires on ESG. The information processing device 100 supports the creation of responses to questionnaires at user companies, for example, by using a RAG (Retrieval Augmented Generation) mechanism, and includes a questionnaire acquisition unit 110, a questionnaire conversion unit 120, a search unit 130, and a response acquisition unit 140. Furthermore, the information processing device 100 includes a corporate information database 150. However, the information processing device 100 is not limited to this configuration; the corporate information database 150 may also be provided by an external device that can communicate with the information processing device 100, such as a server managed by the user company. The information processing device 100 provides users with a service related to the disclosed technology, namely a service that supports the user company in the process of creating responses to questionnaires using the RAG mechanism (hereinafter referred to as the "response creation support service"). The information processing device 100 is any computer, such as a server or a terminal used by a user. Here, the response creation support service may be provided, for example, in the form of SaaS (Software as a Service). In this case, the information processing device 100 is owned or managed by the provider of the response creation support service (hereinafter referred to as the "service provider"). This specification provides details regarding the case where the information processing device 100 relating to the disclosed technology is, for example, a server owned or managed by a service provider, and the information processing device 100 includes a corporate information database 150.

[0017] The information processing device 100 has, in addition to the aforementioned corporate information database 150, an item table (not shown). The item table is a table that describes each ESG evaluation item, the disclosure standards corresponding to those items, the evaluation standards of each evaluation agency corresponding to those items, and the importance of those items in a corresponding manner. The item table is pre-configured by the service provider and stored, for example, in an unillustrated storage unit accessible by the information processing device 100. ESG evaluation items are, for example, created by the service provider. Disclosure standards are, for example, created by the service provider by collecting disclosure standard information publicly available on the internet and analyzing the correspondence between each disclosure standard and the ESG evaluation items. Importance indicates how important each ESG evaluation item is considered among multiple ESG evaluation items, and may be set by the service provider using a default calculation formula, or by the user receiving the response creation support service. The importance values ​​included in the item table are those set by default.

[0018] Corporate Information Database The Corporate Information Database 150 holds corporate information held by companies, including user companies. The corporate information held by the Corporate Information Database 150 includes, for example, information on company names, disclosure information on companies' ESG related to ESG evaluation items, CDP (Credit Default Swap) information, and ESG evaluation information, which is evaluation information of companies by ESG rating agencies. ESG rating agencies are organizations that evaluate whether companies are conducting activities that take ESG into consideration. Furthermore, company information includes, for example, the information disclosed by each company in response to various disclosure systems that require disclosure in accordance with ESG disclosure standards, as information corresponding to each indicator stipulated by those disclosure systems.

[0019] Disclosure Information The corporate information database 150 holds disclosure information from one or more companies. This information is obtained, for example, through crawling. Crawling is a well-known technique in which a program called a crawler periodically visits multiple websites to acquire and store information.

[0020] Here, for example, the information processing device 100 first periodically visits the websites of multiple companies designated in advance by the service provider, thereby acquiring electronic data of disclosure documents from those companies. The information processing device 100 then links the acquired electronic data of disclosure documents from the companies with information that identifies the company disclosing them (hereinafter referred to as "company identification information") and stores it in the company information database 150. Company-specific information includes, for example, the company's official name, abbreviation, or a unique ID or code. Hereafter, company-specific information will be defined as the company's official name or abbreviation. Furthermore, the information processing device 100 may link the acquired electronic data of corporate disclosure documents with information that identifies those disclosure documents (hereinafter referred to as "document identification information") and store them in the corporate information database 150. Document identification information includes, for example, the official name of the disclosed document, an abbreviation of the disclosed document, the type of disclosed document, or an ID or code specific to the type of disclosed document. Hereinafter, the electronic data of the disclosed document will be stored linked to the document identification information, and the document identification information will be the document name, such as the official name or abbreviation of the disclosed document. The storage of acquired electronic data of disclosure documents in the corporate information database 150 may be done in a manner that overwrites the electronic data of disclosure documents already stored for a given company, or it may be done in a manner that adds the newly acquired electronic data of disclosure documents to the electronic data of disclosure documents already stored, for example, after linking it with information on when each electronic data of disclosure documents was acquired.

[0021] The electronic data of the disclosure documents stored in the corporate information database 150 is processed in a way that makes it easy for the information processing device 100 related to this disclosure technology to handle. For example, machine learning techniques may be used to process the electronic data of disclosure documents stored in the corporate information database 150. Specifically, when electronic data of disclosure documents is input, a trained model may be used that is trained to output, for each corresponding section, the content of the section corresponding to one of the subcategories of ESG evaluation items (hereinafter referred to as "corresponding section"), information indicating the location of that corresponding section, information on the directory or link to which the location is accessed (hereinafter referred to as "link destination information"), and a subcategory code that identifies the subcategories of ESG evaluation items to which the corresponding section corresponds. The trained model may be composed of, for example, an artificial neural network, or it may be a large-scale language model. Hereinafter, the information indicating the location of the document and the linked information will also be referred to as "location information." The information indicating the location of the document includes the name of the disclosed document (document name), the page number in the disclosed document where the corresponding section is described, or the paragraph number on the page indicated by that page number (for example, a number indicating which paragraph it is from the top on that page), or the figure or table number, etc.

[0022] This disclosed technology may use a large language model (LLM) from among the trained models as a means of extracting information from the disclosed materials. When a large-scale language model is used, for example, the information processing device 100 inputs electronic data of disclosure documents newly acquired by crawling and stored in the corporate information database 150 into the large-scale language model. The prompts at that time may include, for example, the electronic data of the disclosure documents themselves or information that specifies the electronic data of the disclosure documents, and the information that indicates the ESG evaluation items themselves or information that specifies the item table. Furthermore, the prompts may include instructions to extract the corresponding section and its location corresponding to one of the subcategories of the ESG evaluation items, instructions to generate different links for each location and generate link destination information that transitions to that location, instructions to generate information linking each link to each subcategory code (hereinafter referred to as "item link linking information"), and instructions including specific rules for generating links. Prompts for input into a large-scale language model can be automatically generated. For example, a prompt template is set by the service provider and stored in an unillustrated storage unit accessible by the information processing device 100. When processing the electronic data of the disclosure document, the information processing device 100 can retrieve the template from the aforementioned storage unit and automatically create the prompt described above by embedding necessary information, such as information specifying the electronic data of the disclosure document.

[0023] The large-scale language model may be provided by an external server or by the information processing device 100. For example, an existing large-scale language model provided as a service by an external server may be used. In this case, the information processing device 100 can input prompts to the large-scale language model and obtain responses to the inputs from the large-scale language model through API (Application Programming Interface) communication.

[0024] When the information processing device 100 receives output from a trained model, including corresponding locations, location information, and subcategory codes for the subcategories of ESG evaluation items, it links this information with company identification information and stores it in the company information database 150 as disclosure information. For example, one disclosure document for one company may typically contain multiple pieces of disclosure information. Therefore, the company information database 150 usually stores multiple pieces of disclosure information linked together for a single company. Thus, in this embodiment, the "disclosed information" is identified by including at least the corresponding location, location information, and the classification of the ESG evaluation items (in this case, the sub-classification code). In the example above, the classification of ESG evaluation items is identified by classification codes (sub-classification codes), but the classification of ESG evaluation items may also be identified by classification names and classification codes. Furthermore, the corresponding section may include text, figures, tables, or pages in the disclosed document. The text designated as the corresponding section may be the entire text or a part of it. For example, the text designated as the corresponding section may be identified as a part of a paragraph or as a whole paragraph. The specific corresponding parts of a chart or graph may include all of the charts or graphs themselves, or some of the charts or graphs themselves, for example, if there are multiple charts or graphs. Furthermore, the pages designated as the corresponding sections may include all pages in a given section, or only some of the pages. The corresponding section may include multiple types of elements, such as text, diagrams, or pages. For example, the corresponding section may be a part of text (e.g., a paragraph) and a part of multiple diagrams or tables, a part of a page from a group of pages and a part of a diagram or table, or a specific page from a group of pages and the paragraph on that page.

[0025] Furthermore, location information may include the name of the disclosed document, the page or page and paragraph in the disclosed document that contains the corresponding section, and a link to the page or page and paragraph containing the corresponding section. In this case, pages are identified by their page numbers, and paragraphs can be identified by their page numbers and the paragraph numbers on the page indicated by those page numbers (for example, a number indicating which paragraph it is from the top on that page).

[0026] Furthermore, the processing of the electronic data of the disclosed documents may be done using a pre-trained model as described above, a so-called rule-based algorithm, or manually. Regardless of the processing method used, it is sufficient that the resulting disclosed information is stored in the corporate information database 150.

[0027] Prior to the search described later, the corporate information database 150 pre-stores, in the manner described above, the disclosure information for each company, including the corresponding location, location information, and classification of ESG evaluation items (in this case, sub-classification codes), in an organized manner. Figure 2 shows an example of a data structure for an information processing program related to the disclosed technology. As illustrated in Figure 2, the data should preferably include the following data elements for each piece of disclosed information: "Company name," "Document name," "Text in the document," "Diagrams in the document," "Link information to the content (page, paragraph number, etc.)," ​​"ESG classification," "ESG evaluation item (major category)," "ESG evaluation item (minor category)," and "ESG evaluation item (code)." Of these, the "Company Name," "Document Name," "ESG Classification," "ESG Evaluation Items (Major Classification)," "ESG Evaluation Items (Minor Classification)," and "ESG Evaluation Items (Code)" are as described above. "Text within the document" refers to the content of the text or page itself when the corresponding section of the disclosed information is a document or page. Therefore, this data element stores character data that represents the content of the text or page itself. "Figures included in the document" refers to the content of the figure or table itself when the corresponding section of the disclosed information is a figure or table. Therefore, this data element stores image data or similar data representing the figure or table itself. "Content link information (page number, paragraph number, etc.)" refers to location information. As mentioned above, location information includes information indicating the location of the corresponding section and link destination information. Therefore, this data element stores the link to the disclosed information and information indicating where the disclosed information is located.

[0028] Alternatively, the disclosed information may be stored in the corporate information database 150, for example, as follows: For example, first, the information processing device 100 periodically visits the websites of multiple companies designated in advance by the service provider, thereby acquiring disclosure information from those companies. The information processing device 100 first converts the acquired disclosure information from the companies into text data using optical character recognition (OCR), then divides the converted text data into predetermined ranges, and converts the text data into vectors for each divided range. The information processing device 100 then links the vectors obtained through the conversion with information that identifies the company disclosing the information (company identification information) and stores them in the company information database 150. If the disclosed information is created in PDF format, the disclosed information may already contain text data. In that case, the information processing device 100 may not need to convert the disclosed information into text data using optical character recognition (OCR) as described above. Furthermore, the disclosed information of multiple companies acquired by the information processing device 100 may be created in HTML format. When the information processing device 100 divides the aforementioned text data into predetermined ranges, it can divide it, for example, by page, by paragraph, or by the number of characters. When dividing the text data by the number of characters, the information processing device 100 can set any number of characters, such as a maximum of 400 characters, as the unit of division. In this case, the information processing device 100 may also divide the text data so that there is an overlap of any number of characters, such as 100 characters, before and after the divided range.

[0029] ESG Evaluation Information As mentioned above, ESG evaluation information includes company-specific evaluation information and evaluation criteria information. The raw information (hereinafter referred to as "raw evaluation information") of the evaluations conducted by each evaluation agency for each company, which is necessary for creating company-specific evaluation information among the ESG evaluation information, is uploaded to the information processing device 100, for example, when a company, as a user possessing such raw evaluation information, operates the user terminal UT. In other words, the raw evaluation information is information that shows the evaluations conducted by each evaluation agency for each company before they were divided into separate ESG evaluation items. The process of extracting the evaluations performed by each evaluation agency in the original evaluation information, categorized by ESG evaluation item, that is, the process of extracting company-specific evaluation information (hereinafter referred to as the "extraction process"), may be performed manually or by the information processing device 100 using machine learning technology. The procedure by which the information processing device 100 performs the extraction process is referred to as the "evaluation information extraction procedure". When machine learning techniques are used, specifically, a pre-trained model may be used that, upon input of raw evaluation information, is trained to categorize the evaluations conducted by each evaluation agency according to their respective ESG evaluation items. The pre-trained model may be composed of an artificial neural network, or it may be a large-scale language model (LLM) such as GPT (Generative Pre-trained Transformer).

[0030] The prompts input to the large-scale language model include, for example, the original evaluation information itself or information specifying the original evaluation information, and the information itself that shows the ESG evaluation items or information that specifies the item table. The information specifying the original evaluation information is, for example, information that indicates the directory and file name of the original evaluation information if it is stored in an unillustrated storage unit accessible by the information processing device 100. Similarly, the information specifying the item table is information that indicates the directory and file name of the item table. Furthermore, the prompt includes instructions to extract the corresponding sections from the original evaluation information that contain descriptions related to the ESG evaluation items, and to extract the evaluation results, such as scores, included in those sections. Furthermore, the processing based on these instructions may be carried out in multiple steps using multiple types of prompts. For example, first, the corresponding sections may be extracted using a prompt that only contains instructions for extracting the corresponding sections, and then the evaluation results, such as scores, included in the extracted corresponding sections may be extracted using a prompt that contains instructions for extracting those evaluation results. Furthermore, the prompt may include instructions to extract evaluation criteria information included in the corresponding section. When a large-scale language model is used as the trained model, prompts can be automatically generated. For example, a prompt template is set by the service provider and stored in an unillustrated memory accessible by the information processing device 100. When the information processing device 100 performs the task of separating the evaluations made by each evaluation organization in the original evaluation information into ESG evaluation items, it can retrieve the template from the aforementioned memory and automatically create the prompt described above by embedding necessary information, such as information specifying the original evaluation information.

[0031] The large-scale language model may be provided by an external server or by the information processing device 100. For example, an existing large-scale language model provided as a service by an external server may be used. In this case, the information processing device 100 can input prompts to the large-scale language model and obtain responses to the inputs from the large-scale language model through API integration.

[0032] The extraction process may be performed using a pre-trained model, a so-called rule-based algorithm, or entirely manually, as described above. Furthermore, the extraction process may be performed as a semi-automated operation, where it is initially carried out by a pre-trained model such as the large-scale language model mentioned above, and then finally confirmed by human visual inspection. Furthermore, among the ESG evaluation information, the evaluation criteria information may be extracted in association with ESG evaluation items during the extraction process as described above and stored in the corporate information database 150, or it may be created after the creation of corporate evaluation information by referring to the item table to associate it with ESG evaluation items and then stored in the corporate information database 150. This process can be performed automatically by the information processing device 100 or manually.

[0033] 《Survey Collection Section 110》 The questionnaire acquisition unit 110 acquires questionnaire data from the user terminal UT. The questionnaire data is data that indicates a questionnaire to be answered by the user. The questionnaire includes, for example, one or more question fields and answer fields corresponding to those question fields. The user can operate the user terminal UT to access the information processing device 100 and display a screen for using the response creation support service on the user terminal UT's display (not shown). This screen will be displayed in a browser, for example, if the response creation support service is provided in the form of SaaS. This screen can display, for example, a survey data registration window. The user can operate the user terminal UT to enter information to identify any survey data into this survey data registration window. When the user enters information to identify any survey data into the survey data registration window, the user terminal UT transmits the survey data identified by the information entered by the user to the information processing device 100.

[0034] For example, when a user responds to a questionnaire sent by Company A, a business partner of the user's company, the user operates the user terminal UT and enters information indicating the storage location of the questionnaire data sent by Company A into the questionnaire data registration window. When the user enters the above information into the questionnaire data registration window, the user terminal UT transmits the questionnaire data stored at the location indicated by the above information to the information processing device 100. The questionnaire acquisition unit 110 of the information processing device 100 acquires questionnaire data transmitted from the user terminal UT. Once the questionnaire acquisition unit 110 acquires the questionnaire data transmitted from the user terminal UT, it outputs the acquired questionnaire data to the questionnaire conversion unit 120. The questionnaire data also includes information that identifies the company to which the user who submitted the questionnaire belongs (company identification information).

[0035] 《Questionnaire Conversion Section 120》 The questionnaire conversion unit 120 obtains questionnaire data from the questionnaire acquisition unit 110. When the questionnaire conversion unit 120 obtains questionnaire data from the questionnaire acquisition unit 110, it uses the obtained questionnaire data to obtain data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format (hereinafter referred to as "modified questionnaire data"). Specifically, first, the survey conversion unit 120 checks the data format of the acquired survey data. If the data format of the acquired survey data is not a predetermined data format, the survey conversion unit 120 analyzes the survey data and obtains the analysis results. Here, the predetermined data format is a data format that can be processed by the first machine learning model into which the survey data is input. The first machine learning model will be described later. Here, the predetermined data format is assumed to be, for example, PDF (Portable Document Format).

[0036] The questionnaire conversion unit 120 checks the data format of the acquired questionnaire data, and if the data format of the questionnaire data is not PDF, it performs an analysis such as the following. For example, if the data format of the survey data is a table format created with spreadsheet software such as Excel, the survey conversion unit 120 analyzes the survey data using a known library for analyzing the information of cells contained in the table (e.g., openpyxl), and extracts as analysis results information about the arrangement of the question fields and answer fields included in the survey, as well as information about the content of the questions included in the survey (e.g., question text). Furthermore, if the data format of the survey data is a document format created with word processing software such as Word, the survey conversion unit 120 analyzes the survey data using a known library, similar to the case where the data format of the survey data is a tabular format, and extracts information about the content of the questions included in the survey as an analysis result. Furthermore, if the data format of the survey data is not PDF, which is a predetermined data format, and the survey data contains image data, the survey conversion unit 120 converts the data format of the survey data to PDF, which is a predetermined data format. For example, if the data format of the survey data is an office format including a table format created with spreadsheet software such as Excel, or a document format created with word processing software such as Word, and the survey data contains image data, the survey conversion unit 120 uses known data format conversion software to convert the data format of the survey data to PDF, which is a predetermined data format.

[0037] Next, the questionnaire conversion unit 120 generates a prompt to input to the first machine learning model (hereinafter referred to as the "modification prompt") which instructs the system to modify the format of the questionnaire identified by the acquired questionnaire data. Specifically, the questionnaire conversion unit 120 generates a modification prompt which instructs the system to modify the format of the questionnaire identified by the questionnaire data into a format that can be processed by the second machine learning model, which will be described later. For the sake of simplicity, in the following explanation, the questionnaire after its format has been modified will be referred to as the "modified questionnaire."

[0038] Here, the modified prompt includes, for example, the following: (1) Questionnaire data obtained by the questionnaire acquisition unit 110 (2) When the data format of the above-mentioned questionnaire data is not PDF, the results of the analysis of the questionnaire data by the questionnaire conversion unit 120. (3) Instructions to identify the answer fields in the questionnaire identified by the above questionnaire data, and the question fields corresponding to those answer fields. (4) An instruction to identify the category of each question included in the questionnaire, which is identified from the above questionnaire data, and to output the identified category. (5) An instruction to identify the answer format for each question included in the questionnaire identified by the above questionnaire data, and to output the identified answer format. (6) Instructions to output data showing the modified questionnaire so that the number of question fields and answer fields in the modified questionnaire are equal.

[0039] The questionnaire conversion unit 120 inputs a modified prompt containing the contents of (1) to (6) above into the first machine learning model. The questionnaire conversion unit 120 may also input at least one of (1) or (2) above directly into the first machine learning model without including it in the modified prompt. Furthermore, if the data format of the questionnaire data acquired from the questionnaire acquisition unit 110 is a predetermined data format (in this case, PDF), the questionnaire conversion unit 120 may omit the analysis of the questionnaire data described above. In that case, the questionnaire conversion unit 120 may generate a modified prompt that includes at least the contents of (3) to (6), excluding the results of the analysis of the questionnaire data exemplified in (2) above, and input the generated modified prompt into the first machine learning model. This is because if the data format of the questionnaire data is a predetermined data format, the first machine learning model can analyze the questionnaire data as is. In other words, the first machine learning model is pre-trained to handle questionnaire data created in a predetermined data format (in this case, PDF). On the other hand, if the data format of the questionnaire data is not a predetermined data format, the questionnaire conversion unit 120 may perform the analysis of the questionnaire data as described above and input the analysis results into the first machine learning model.

[0040] The first machine learning model may be composed of, for example, an artificial neural network, and there may be multiple such models. Furthermore, if there are multiple first machine learning models, the first machine learning model may include a large-scale language model (LLM). In this case, the first machine learning model is a machine learning model that, when given the modified prompt described above as input, outputs data representing a modified questionnaire (hereinafter referred to as "modified questionnaire data") as a response.

[0041] The first machine learning model, which includes a large-scale language model, may be provided by an external server or by the information processing device 100. Furthermore, as the large-scale language model, for example, an existing large-scale language model provided as a service by an external server may be used. In this case, the information processing device 100 can input modification prompts to the large-scale language model and obtain responses to the input from the large-scale language model via API integration. The first machine learning model generates modified questionnaire data based on a given modification prompt and outputs the generated modified questionnaire data to the information processing device 100. The questionnaire conversion unit 120 of the information processing device 100 acquires the modified questionnaire data output from the first machine learning model. Once the questionnaire conversion unit 120 acquires the modified questionnaire data from the first machine learning model, it outputs the acquired modified questionnaire data to the search unit 130 and the response acquisition unit 140.

[0042] Figure 3 shows an example of a questionnaire identified by the questionnaire data. Figure 4 shows an example of a modified questionnaire identified by the modified questionnaire data. For example, as shown in Figure 3, a questionnaire consists of multiple question fields in which questions are written, an answer field corresponding to each question field, a text field for writing the answer, a field for entering the department of the person in charge of answering, and a field for entering the name of the person in charge. A modified questionnaire also consists of, for example, a field indicating the question identifier (coordinate), a question number field (question_number), a field describing the category of the question (question_category), a field describing the question text (question_text), and a field describing answer options (answer_options). The first machine learning model, upon receiving survey data showing a questionnaire as shown in Figure 3, along with the modified prompt described above, outputs modified survey data showing a modified questionnaire as shown in Figure 4. Although not shown in Figure 4, in a modified questionnaire as shown in Figure 4, a response field for entering answers may be provided to the right of the field describing the answer options, and a field for describing the reason for the answer may be provided further to the right of that. Furthermore, in the modified questionnaire shown in Figure 4, for example, a field describing the answer format may be provided to the right of the field in which the question text is written. The answer format indicates, for example, how the user will answer, and may include a selection format using radio buttons or a multiple-choice format using checkboxes.

[0043] The first machine learning model, upon receiving survey data and a modified prompt as input, should generate modified survey data according to the modified prompt, such that the number of fields containing question texts and the number of answer fields for filling in answers are equal. In this case, it is assumed that the number of answer fields in the survey identified by the survey data is equal to the total number of questions in that survey. Furthermore, if the first machine learning model analyzes the input survey data and finds that a single question field contains multiple question sentences, the modified survey may divide the question field and answer field for each of these question sentences, thereby generating modified survey data in which the question field and answer field are divided for each question sentence. The questionnaire shown in Figure 3 and the modified questionnaire shown in Figure 4 are merely examples, and the questionnaire and modified questionnaire may be generated with content and format other than those described above. However, it is desirable that the modified questionnaire be in a format that allows the second machine learning model, described later, to accept and process the modified questionnaire data when it is input into the second machine learning model.

[0044] Search section 130 The search unit 130 retrieves the modified questionnaire data from the questionnaire conversion unit 120. Once the search unit 130 retrieves the modified questionnaire data from the questionnaire conversion unit 120, it retrieves company information from the company information database 150 to answer the questions included in the modified questionnaire. For example, the search unit 130 obtains information indicating the questions included in the modified questionnaire (hereinafter referred to as "question information") based on the acquired modified questionnaire data, and generates a search query to search for company information necessary to answer the questions included in the modified questionnaire based on the acquired question information. Here, the question information includes, for example, text data describing the question and text data describing the category of the question. The search unit 130 can obtain the text data describing the question from the field in which the question is described (question_text), and can obtain the text data describing the category of the question from the field in which the category of the question is described (question_category). When the search unit 130 generates a search query, it performs a vector search of the company information database 150 based on the generated search query and obtains the company information necessary to answer the modified questionnaire as a search result.

[0045] Furthermore, the search unit 130 may arbitrarily set an upper limit on the number of company information entries to retrieve from the company information database 150 based on the generated search query, as a search condition when searching the company information database 150. For example, the search unit 130 may set an upper limit on the number of company information entries to retrieve as a search condition, such as retrieving up to the top 10 company information entries that have a high percentage of matching the search query as search results. Here, the company information acquired by the search unit 130 includes, for example, information about the company name, disclosure information disclosed by the company, CDP information, and evaluation information of the company by ESG rating agencies. In addition, each of these pieces of information included in the company information may be assigned a predetermined tag. In that case, the search unit 130 may acquire company information that matches the search query based on the tag information. The search unit 130 outputs the acquired company information to the response acquisition unit 140. For the sake of simplicity, in the following explanation, the company information acquired from the company information database 150 that is necessary to answer the modified questionnaire will be referred to as "specific company information."

[0046] 《Answer acquisition part 140》 The response acquisition unit 140 acquires modified survey data from the survey conversion unit 120. The response acquisition unit 140 also acquires specific company information from the search unit 130. The response acquisition unit 140 acquires modified questionnaire data from the questionnaire conversion unit 120 and specific company information from the search unit 130. Using the acquired specific company information and the modified questionnaire data, it acquires data indicating the answers to the questions included in the modified questionnaire. Specifically, the response acquisition unit 140 generates a prompt to be input to the second machine learning model (hereinafter referred to as the "response prompt"), which includes an instruction to generate an answer to a question included in the acquired modified questionnaire. At this time, the response acquisition unit 140 may also include an instruction in the response prompt to have the second machine learning model output the reason for the generated answer. Furthermore, if the response format of a question included in the modified questionnaire is multiple-choice, that is, if the response format is such that the respondent is required to select the appropriate option from among several options, the response acquisition unit 140 may include information about those multiple options in the response prompt.

[0047] When the response acquisition unit 140 generates a response prompt, it inputs the generated response prompt, the modified questionnaire data obtained from the questionnaire conversion unit 120, and the specific company information obtained from the search unit 130 into the second machine learning model. The response acquisition unit 140 may input the response prompt, the modified questionnaire data, and the specific company information into the second machine learning model individually, or, if the modified questionnaire data and the specific company information can be included in the response prompt, it may generate the response prompt including the modified questionnaire data and the specific company information, and then input the generated response prompt into the second machine learning model.

[0048] The second machine learning model may be composed of, for example, an artificial neural network, and there may be multiple such models. Furthermore, if there are multiple second machine learning models, they may include large-scale language models (LLMs). In this case, the second machine learning model is a machine learning model trained to output data indicating the answers to the questions included in the modified questionnaire (hereinafter referred to as "answer data") when given the above-mentioned answer prompts, modified questionnaire data, and specific company information as input. Alternatively, the second machine learning model may be a machine learning model trained to output data indicating the reasons for the answers (hereinafter referred to as "answer reason data") in addition to the above-mentioned answer data when given the above-mentioned answer prompts, modified questionnaire data, and specific company information as input.

[0049] The second machine learning model, which includes a large-scale language model, may be provided by an external server or by the information processing device 100. Furthermore, as the large-scale language model, for example, an existing large-scale language model provided as a service by an external server may be used. In this case, the information processing device 100 can input response prompts, modified survey data, and specific company information to the large-scale language model via API integration, and can also obtain responses to the input from the large-scale language model. The second machine learning model generates response data, or response data and response reason data, based on the given response prompt, modified questionnaire data, and specific company information, and outputs the generated data to the information processing device 100. The response acquisition unit 140 of the information processing device 100 acquires the response data, or response data and response reason data, output from the second machine learning model.

[0050] In the above explanation, an example was described in which the response acquisition unit 140 acquires response data, or response data and response reason data, from the second machine learning model. However, the response acquisition unit 140 may also acquire modified questionnaire data in which these responses, or responses and reasons for responses, have been entered into the modified questionnaire, from the second machine learning model. In other words, the response acquisition unit 140 may acquire modified questionnaire data, including response data and response reason data, from the second machine learning model. In that case, the response acquisition unit 140 only needs to include an instruction in the response prompt given to the second machine learning model to output modified questionnaire data in which responses have been entered into the response field, or modified questionnaire data in which responses have been entered into the response field and reasons for responses have been entered into the reason field. Furthermore, the first machine learning model and the second machine learning model described above may be the same machine learning model, and this same machine learning model may be the same large-scale language model (LLM). Furthermore, in the above example, the search unit 130 may perform the search using a technique called "few shots." For example, the search unit 130 generates only company names as a search query and retrieves company information matching the search query from the company information database 150. The response retrieval unit 140 may then include all the company information retrieved by the search unit 130 in the response prompt and prompt the second machine learning model to generate response data, or response data and response reason data. In this case, the information processing device 100 can obtain optimal response data, etc., using only a small amount of information (in this case, company names).

[0051] Figure 5 shows an example of a modified questionnaire with input for answers and reasons for answers. As shown in Figure 5, the modified questionnaire is composed of fields such as a question identifier (question_id), a question number (question_number), a question category (question_category), a question text (question_sentence), an answer format (answer_format), answer options (options), an answer field (answer), and a reason field (reason) for explaining the reason for the answer. Also, as shown in Figure 5, the answer field contains the answer to the question, and the reason field contains the reason for the answer. The reason for the answer, as shown in Figure 5 for example, describes which page of the company information the reason for the answer entered in the answer field is found on, or in other words, which page of the company information is the basis for the answer. For example, the first line, which has "1" as the question identifier, has the question category described as "1. Environment (E) Area / 1 GHG Initiatives (Energy)" and the question text is "Is your company taking steps to reduce energy consumption?". The answer format for this question is a radio button selection, with the options being "Yes" or "No". The answer to this question is "Yes", and the reason given is "The Corporate Governance Report 2023 (p.12) states that the company is promoting the expansion of renewable energy use and energy conservation activities." This reason is generated based on specific company information obtained by the search unit 130. Furthermore, in lines 2 through 5, where the question identifier is displayed as "2.1" to "2.4", the answer choices for the same question are displayed separately. Specifically, in lines 2 through 5, the question category is described as "1. Environment (E) area / 1. Regarding GHG initiatives (energy)", and the question is described as "Q2 is for business partners who answered Yes to Q1. What initiatives are you undertaking? Please select all that apply." The answer format for this question is a multiple-choice format using checkboxes, and the choices are "1. We have set a total CO2 target within the company", "2. We are using PDCA management to achieve the total target", "3. We have publicly announced our commitment to carbon neutrality by 2050", and "4. We are building a company-wide cross-functional system to strengthen our medium- to long-term environmental initiatives". In addition, the answer to whether or not to select any of these options is "Yes", and the reason for selecting each option is described. These reasons are also generated based on specific company information obtained by the search unit 130.

[0052] The reasons for the answers shown in Figure 5 are merely examples, and any content that clearly explains the reason for the answer does not necessarily have to be written in text. Furthermore, the reason field for describing the reason for the answer, as exemplified in Figure 5, is not a mandatory feature in the modified questionnaire and may be omitted. However, including a reason field in the modified questionnaire allows users to easily understand the reason for their answer. Therefore, it is preferable to include a reason field in the modified questionnaire. Furthermore, the example in Figure 5 illustrates a case where the second machine learning model outputs modified questionnaire data in which the answers and reasons for the answers have been input into the modified questionnaire. However, the second machine learning model may not output modified questionnaire data after inputting the answers and reasons for the answers into the modified questionnaire, but rather output only the answer data, or only the answer data and reason data, after associating them with the question identifier or question number. In this case, the response acquisition unit 140 acquires only the response data, or only the response data and the reason for the response, from the second machine learning model, but the acquired response data, or the response data and the reason for the response, should be integrated into the modified questionnaire data. In other words, the response, or the response and the reason for the response, should be entered into the modified questionnaire. In this case, the data integration may be performed manually by, for example, the provider of the response creation support service (service provider), or it may be performed by a data integration unit (not shown) provided in the information processing device 100. Furthermore, the response acquisition unit 140 may output response data, or modified questionnaire data which integrates response data and response reason data, to the user terminal UT. In this case, the user terminal UT acquires the modified questionnaire data from the response acquisition unit 140. Once the user terminal UT acquires the modified questionnaire data from the response acquisition unit 140, it displays the modified questionnaire identified by the acquired data on the display of the user terminal UT. For example, the user terminal UT displays the modified questionnaire on the screen used to access the response creation support service, which is displayed on the display.

[0053] Next, we will explain the operation of the information processing device 100 in the information processing system and the flow of each piece of information, referring to Figure 6. In the following explanation, it is assumed that the user company has already received survey data from multiple business partners. Furthermore, in the following explanation, it is assumed that the service provider and the information processing device 100 have already constructed a corporate information database 150. Also, it is assumed that the first machine learning model and the second machine learning model described above are different machine learning models. First, the user operates the user terminal UT to input information to identify the desired survey data into the survey data registration window described above. Once the user has entered the above information into the survey data registration window, the user terminal UT transmits the survey data identified by the input information to the information processing device 100 (indicated by (1) in Figure 6). Next, in the information processing device 100, the questionnaire acquisition unit 110 acquires the questionnaire data transmitted from the user terminal UT (indicated by (2) in Figure 6). Once the questionnaire acquisition unit 110 acquires the questionnaire data from the user terminal UT, it outputs the acquired questionnaire data to the questionnaire conversion unit 120. Next, the questionnaire conversion unit 120 acquires questionnaire data from the questionnaire acquisition unit 110. Using the questionnaire data acquired from the questionnaire acquisition unit 110, the questionnaire conversion unit 120 acquires data indicating a modified questionnaire (modified questionnaire data) in which the format of the questionnaire identified by the said data has been modified to a predetermined format. Specifically, the questionnaire conversion unit 120 generates the modified prompt described above and inputs the generated modified prompt to the first machine learning model 160 (indicated as (3) in Figure 6). The first machine learning model 160 generates modified questionnaire data based on the input modification prompt and outputs the generated modified questionnaire data to the information processing device 100 (indicated by (4) in Figure 6). The questionnaire conversion unit 120 of the information processing device 100 acquires the modified questionnaire data output from the first machine learning model. Once the questionnaire conversion unit 120 acquires the modified questionnaire data from the first machine learning model, it outputs the acquired modified questionnaire data to the search unit 130 and the response acquisition unit 140. Next, the search unit 130 obtains the modified questionnaire data from the questionnaire conversion unit 120 and retrieves company information (specific company information) from the company information database 150 for answering the questions included in the modified questionnaire (indicated as (5) in Figure 6). The search unit 130 outputs the retrieved specific company information to the response retrieval unit 140. Next, the response acquisition unit 140 acquires modified questionnaire data from the questionnaire conversion unit 120 and specific company information from the search unit 130. The response acquisition unit 140 uses the acquired specific company information and modified questionnaire data to acquire data indicating the answers to the questions included in the modified questionnaire. Specifically, the response acquisition unit 140 generates a response prompt that includes an instruction to generate an answer to a question included in the acquired modified questionnaire, and inputs the generated response prompt to the second machine learning model 170 (indicated as (6) in Figure 6). The second machine learning model 170 generates response data based on the input response prompt and outputs the generated response data to the information processing device 100 (indicated by (7) in Figure 6). At this time, the second machine learning model 170 may also generate response reason data along with the response data and output these generated data to the information processing device 100. Alternatively, the second machine learning model 170 may include the response data, or the response data and response reason data, in the modified questionnaire data and output it to the information processing device 100. The response acquisition unit 140 acquires the data output from the second machine learning model 170. Although not shown in Figure 6, the response acquisition unit 140 may also output the data acquired from the second machine learning model 170 to the user terminal UT. The information processing device 100 operates as described by codes (1) to (7) above, and each data is transmitted and received. Here, the RAG mechanism is realized by codes (3) to (7).

[0054] In the example above, the information processing device 100 is shown to perform one instance of sending and receiving information with the second machine learning model 170. However, if the information processing device 100 uses a large-scale language model as the second machine learning model 170, it may perform multiple instances of sending and receiving information with the large-scale language model (multi-turn). For example, in the first turn, the response acquisition unit 140 of the information processing device 100 generates a response prompt as described above and inputs the generated response prompt to the large-scale language model. The response acquisition unit 140 then acquires response data from the large-scale language model. In the second turn, the response acquisition unit 140 generates a prompt for the large-scale language model that asks about the probability or success of the answer indicated by the response data generated in the first turn, and inputs the generated prompt into the large-scale language model. The response acquisition unit 140 then acquires the information regarding the probability or success of the answer output from the large-scale language model and outputs the acquired information to the user terminal UT. In this case, the user not only finds it easier to create answers to questions, but can also easily understand the probability or success of the answer, and can modify the answer as appropriate according to the probability or success of the answer.

[0055] Furthermore, when the response acquisition unit 140 outputs response data obtained from the second machine learning model 170 to the user terminal UT, it may output multiple response data as response candidates to the user terminal UT. For example, after outputting the response data to the user terminal UT, the information processing device 100 collects feedback from the user regarding the response data and stores information about the collected feedback as a log in a storage unit (not shown). User feedback may include, for example, corrected response data or other response data that could be considered for the response data. Then, when the response acquisition unit 140 outputs the response data obtained from the second machine learning model 170 to the user terminal UT, it refers to the log stored in the memory unit, generates multiple response data based on the log, and outputs these generated response data to the user terminal UT as response candidates. In this case, the information processing device 100 can present multiple response candidates to the user at once, and can reduce the number of data transmissions and receptions with the user terminal UT, leading to a reduction in communication volume. Furthermore, the information processing device 100 may also include a function for user-defined response corrections. For example, the information processing device 100 may accept corrections to the response data from the user via the user terminal UT before outputting the response data obtained from the second machine learning model 170 to the user terminal UT via the response acquisition unit 140. The response acquisition unit 140 may then output the user-defined response data to the user terminal UT. In this case, the user can correct the response data in advance on the information processing device 100 before acquiring the response data, leading to improved efficiency in the response creation process. Furthermore, if the information processing device 100 receives a request from the user to modify the response data, it may use the modified response data as retraining data for the second machine learning model 170. In this case, the second machine learning model 170 will be able to output more accurate response data that better reflects the user's intentions, and the information processing device 100 can obtain more accurate response data that better reflects the user's intentions from the second machine learning model 170.

[0056] Furthermore, the above description described an example in which the search unit 130 obtains information indicating the questions included in the modified questionnaire (question information) based on the modified questionnaire data obtained from the questionnaire conversion unit 120, and generates a search query for searching for company information necessary to answer the questions included in the modified questionnaire based on the obtained question information. However, the search unit 130 is not limited to this, and may, for example, obtain question information based on questionnaire data obtained by the questionnaire acquisition unit 110, and generate a search query based on the obtained question information. In other words, the search unit 130 may generate a search query based on data indicating the questionnaire before the format was modified, rather than data indicating the modified questionnaire after the format was modified. Furthermore, the above explanation described an example where the search unit 130 searches for company information stored in the company information database 150. However, the search unit 130 is not limited to this, and may also search for, for example, response data from surveys previously answered by user companies, or response data and response reason data (hereinafter referred to as "answered data"). In this case, it is preferable that the answered data be stored in the company information database 150 in advance, either as part of the company information or as separate information from the company information. If the corporate information database 150 is included in the information processing device 100, the answered data can be uploaded to the information processing device 100 and stored in the corporate information database 150, for example, when the user operates the user terminal UT. Alternatively, if the corporate information database 150 is located on a server managed by the user company, the answered data can be stored in the corporate information database 150 located on that server, for example, when the user operates the user terminal UT. If the completed response data is stored in the company information database 150 as separate information from the company information, the search unit 130 outputs the company information (specific company information) and completed response data obtained from the company information database 150 to the response acquisition unit 140. When the response acquisition unit 140 generates the above-mentioned response prompt, it inputs the generated response prompt, the modified questionnaire data obtained from the questionnaire conversion unit 120, and the specific company information and completed response data obtained from the search unit 130 into the second machine learning model. The response acquisition unit 140 may input the response prompt, the modified questionnaire data, the specific company information, and the completed response data into the second machine learning model individually, or, if the modified questionnaire data, specific company information, and completed response data can be included in the response prompt, it may generate the response prompt including the modified questionnaire data, specific company information, and completed response data, and input the generated response prompt into the second machine learning model. Furthermore, the response acquisition unit 140 may include an instruction in the response prompt to output response data, or response data and response reason data, to the second machine learning model, based on the previously answered data. This increases the likelihood that the second machine learning model will output response data, or response data and response reason data, that reflects the content of the previously answered data, and the response acquisition unit 140 can acquire response data, or response data and response reason data, that reflects the content of responses previously provided by the user company.

[0057] Furthermore, the above description illustrates an example in which the search unit 130 generates its own search query, searches the corporate information database 150 based on the generated search query, and obtains the corporate information necessary to answer the modified questionnaire. However, the search unit 130 is not limited to this example; for example, it may instruct a large-scale language model (LLM) (not shown) to generate a search query and retrieve corporate information. For example, when the search unit 130 obtains modified questionnaire data from the questionnaire conversion unit 120, it generates a prompt (hereinafter referred to as the "search prompt") that includes an instruction to obtain company information necessary to answer the modified questionnaire from the company information database 150, and inputs this prompt into the large-scale language model described above. When the search unit 130 generates a search prompt, it inputs the generated search prompt and the modified questionnaire data obtained from the questionnaire conversion unit 120 into the large-scale language model described above. The search unit 130 may input the search prompt and the modified questionnaire data into the large-scale language model individually, or, if the modified questionnaire data can be included in the search prompt, it may generate the search prompt including the modified questionnaire data and input the generated search prompt into the large-scale language model. The large-scale language model generates a search query to find company information necessary to answer a modified questionnaire, for example, in accordance with the search prompt. Based on the generated search query, it performs a vector search on the company information database 150 and retrieves the company information necessary to answer the modified questionnaire as a search result. The search unit 130 then outputs the company information retrieved by the large-scale language model as specific company information to the response retrieval unit 140. In this case, the search unit 130 does not perform the search process itself, but rather has the large-scale language model perform the search process, which is expected to reduce the processing load compared to when the search unit performs the search process itself.

[0058] Furthermore, the above description described an example in which the search unit 130 generates a search query itself, searches the corporate information database 150 based on the generated search query, and obtains the corporate information necessary to answer the modified questionnaire, or in which the search unit 130 instructs the large-scale language model to perform these processes. However, it is not necessarily required for the information processing device 100 to perform search processing by the search unit 130 or the large-scale language model. For example, the information processing device 100 can pre-acquire company information from the company information database 150 that is held by a specified company and is used to answer questions included in a modified questionnaire identified by data acquired by the questionnaire conversion unit 120. Furthermore, if the response acquisition unit 140 can acquire response data using the information that has been pre-acquired, then the search unit 130 or the search processing by the large-scale language model becomes unnecessary. In this case, the search unit 130 may be omitted from the information processing device 100.

[0059] Furthermore, the above description illustrates an example in which the user inputs information to identify arbitrary survey data into the survey data registration window displayed on the user terminal UT's screen. Alternatively, the user may specify survey data by operating the user terminal UT, for example, by selecting arbitrary survey data from a pull-down menu displayed in the survey data registration window. When the user selects survey data from the pull-down menu, the user terminal UT transmits the selected survey data to the information processing device 100. The pull-down menu displays a list containing multiple survey data when the user presses a designated location. For example, it could display multiple survey data previously acquired by the user terminal (UT) as options.

[0060] Furthermore, in some cases, as indicated by the symbol (5) in Figure 6, the search unit 130 may be unable to obtain specific company information necessary to answer the question from the company information database 150, or it may be able to obtain the information but the amount of information is insufficient. In such cases, if the second machine learning model 170 is a large-scale language model, the answer acquisition unit 140 may include an instruction in the answer prompt to have the large-scale language model output a message to the company user such as, "The answer may be insufficient because there is insufficient company information necessary to answer the question. Please provide input to supplement the answer as needed," and input this answer prompt to the large-scale language model. Furthermore, if the second machine learning model 170 is a large-scale language model, and the above-mentioned response prompt is input to the large-scale language model, the large-scale language model may include information in the response data that indicates a message such as, "The answer may be insufficient because the necessary company information to answer the question is missing. Please provide additional input to supplement the answer as needed."

[0061] Furthermore, when the response acquisition unit 140 outputs the response data acquired from the second machine learning model 170 to the user terminal UT, if the response format of the response data is not a predetermined response format, the unit may modify the response format so that the response format of the response data acquired from the second machine learning model 170 matches the predetermined response format, and then output the response data to the user terminal UT. The predetermined response format may be, for example, the response format described in the field (answer_format) where the response format is described in the modified questionnaire. Furthermore, when the response acquisition unit 140 outputs the response data acquired from the second machine learning model 170 to the user terminal UT, it may output the response data in a manner that allows the user to input (edit) the response from the user terminal UT. This is because it is not possible to know whether the response indicated by the response data is necessarily the response the user desires, and therefore it leaves room for the user to input (edit) the response.

[0062] In this case, if the user inputs any information regarding the answer using the user terminal UT, the user terminal UT should transmit information indicating the content of the input regarding the answer (hereinafter referred to as "additional input information") to the information processing device 100. In the information processing device 100, an additional input information acquisition unit (not shown) acquires additional input information transmitted from the user terminal UT and stores the acquired additional input information in, for example, an additional input information database (not shown). The additional input information database may be provided by the information processing device 100 or by an external device that can communicate with the information processing device 100. The additional input information stored in the additional input information database is used by the answer acquisition unit 140 when outputting answer data for the same or similar questions related to the said additional input information in subsequent instances. For example, when the response acquisition unit 140 outputs response data acquired from the second machine learning model 170 to the user terminal UT, it refers to the additional input information database. If additional input information related to the response data to be output is stored in the additional input information database, it reflects the content of the additional input information in the response data to be output before outputting the response data to the user terminal UT. For example, if additional input information related to the response data to be output is stored in the additional input information database, the response acquisition unit 140 may modify the response data to be output based on the additional input information stored in the additional input information database and output the modified response data, or it may output the additional input information stored in the additional input information database together with the response data to be output. This allows the response acquisition unit 140 to present the user with an answer that is closer to the answer the user desires.

[0063] Furthermore, the above processing also applies when the response data includes information indicating a message such as, "The response may be insufficient because the necessary company information to answer the question is missing. Please provide additional information to supplement the response as needed." In this case, it is expected that the user will see the above message on the user terminal UT and, if necessary, use the user terminal UT to provide some input to the response. Even in that case, the user terminal UT will send information indicating the content entered for the response (additional input information) to the information processing device 100, and the information processing device 100 will have an additional input information acquisition unit (not shown) acquire the additional input information sent from the user terminal UT and save the acquired additional input information to an additional input information database (not shown). Subsequently, the response acquisition unit 140 will use the additional input information when outputting response data in the same manner as above, thereby being able to present the user with a response that is closer to the response the user desires.

[0064] Embodiment 2. Embodiment 2 shows the implementation of the disclosed technology as an information processing program. Unless otherwise specified, the same reference numerals used in Embodiment 1 are used in Embodiment 2. In Embodiment 2, explanations that overlap with those in Embodiment 1 are omitted as appropriate. The information processing program of Embodiment 2 causes the computer to execute the following: a questionnaire acquisition procedure for acquiring questionnaire data; a questionnaire conversion procedure for acquiring data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format, using the questionnaire data acquired in the questionnaire acquisition procedure; a search procedure for acquiring company information for answering questions included in the modified questionnaire from a database that stores company information held by a predetermined company; and an answer acquisition procedure for acquiring data indicating answers to questions included in the modified questionnaire, using the company information acquired in the search procedure and the data indicating the modified questionnaire acquired in the questionnaire conversion procedure.

[0065] Figure 7 shows the hardware configuration 200 of the information processing device 100 according to the disclosed technology. As shown in Figure 7, the hardware configuration 200 of the information processing device 100 includes a communication interface 210, an input / output interface 220, a processor 230, and a memory 240.

[0066] Figure 8 is a flowchart showing the information processing method implemented by the information processing program relating to the disclosed technology as a processing procedure. As shown in Figure 8, the information processing method implemented by the information processing program includes a questionnaire acquisition procedure ST110, a questionnaire conversion procedure ST120, a search procedure ST130, and a response acquisition procedure ST140. The number assigned to each processing step corresponds to the code of the entity that executes that processing step. For example, the questionnaire acquisition procedure ST110 is executed by the questionnaire acquisition unit 110, the questionnaire conversion procedure ST120 is executed by the questionnaire conversion unit 120, the search procedure ST130 is executed by the search unit 130, and the response acquisition procedure ST140 is executed by the response acquisition unit 140. As shown in Figure 8, the questionnaire acquisition procedure ST110, the questionnaire conversion procedure ST120, the search procedure ST130, and the response acquisition procedure ST140 are all contained within a loop.

[0067] The information processing method implemented by the information processing program according to Embodiment 2 is an information processing method executed by a computer. In this information processing method, when the loop is started, first, the questionnaire acquisition unit 110 executes a questionnaire acquisition procedure (ST110) to acquire questionnaire data transmitted from the user terminal UT. Then, the questionnaire conversion unit 120 uses the questionnaire data acquired in the questionnaire acquisition procedure to execute a questionnaire conversion procedure (ST120) to acquire data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format. Next, the search unit 130 executes a search procedure (ST130) to acquire company information for answering questions included in the modified questionnaire from a company information database 150 that stores company information held by a predetermined company. Furthermore, the answer acquisition unit 140 uses the company information acquired in the search procedure and the data indicating the modified questionnaire acquired in the questionnaire conversion procedure to execute an answer acquisition procedure (ST140) to acquire data indicating answers to questions included in the modified questionnaire. Note that Figure 6 shows an example of loop processing, but the disclosed technology is not limited to this.

[0068] The functions of the questionnaire acquisition unit 110, questionnaire conversion unit 120, search unit 130, and response acquisition unit 140 of the information processing device 100 according to the disclosed technology are realized by processing circuits. That is, the information processing device 100 includes processing circuits for executing the questionnaire acquisition procedure ST110, the questionnaire conversion procedure ST120, the search procedure ST130, and the response acquisition procedure ST140. The processing circuit is a processor 230 (also called a CPU, central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, or DSP) that executes an information processing program stored in the memory 240.

[0069] The functions of the questionnaire acquisition unit 110, the questionnaire conversion unit 120, the search unit 130, and the response acquisition unit 140 are realized by software, firmware, or a combination of software and firmware. The software and firmware are written as information processing programs and stored in the memory 240. The processing circuit realizes the functions of each unit by reading and executing the information processing programs stored in the memory 240. In other words, the information processing device 100 includes a memory 240 for storing information processing programs that, when executed by the processing circuit, will result in the execution of the questionnaire acquisition procedure ST110, the questionnaire conversion procedure ST120, the search procedure ST130, and the response acquisition procedure ST140. It can also be said that these information processing programs cause the computer to execute the procedures or methods of the questionnaire acquisition unit 110, the questionnaire conversion unit 120, the search unit 130, and the response acquisition unit 140. Here, the memory 240 may be, for example, a non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, or EPROM. Furthermore, the memory 240 may be configured to include a disk such as a magnetic disk, flexible disk, optical disk, compact disk, minidisc, or DVD. In addition, the memory 240 may be configured as an HDD or SSD.

[0070] In the above explanation, the search unit 130 described an example where the target of the search is company information stored in the company information database 150. However, the search unit 130 is not limited to this, and may also search the aforementioned answered data, for example, as described in Embodiment 1. In this case, it is preferable that the answered data be stored in the company information database 150 in advance, either as part of the company information or as separate information, using the procedure described in Embodiment 1. In Embodiment 2, as in Embodiment 1, when the response acquisition unit 140 generates a response prompt, it may input the generated response prompt, the modified questionnaire data obtained from the questionnaire conversion unit 120, and the specific company information and completed response data obtained from the search unit 130 into the second machine learning model. Furthermore, the response acquisition unit 140 may include an instruction in the response prompt to output response data, or response data and response reason data, to the second machine learning model, based on the previously answered data. This increases the likelihood that the second machine learning model will output response data, or response data and response reason data, that reflects the content of the previously answered data, and the response acquisition unit 140 can acquire response data, or response data and response reason data, that reflects the content of responses previously provided by the user company.

[0071] Furthermore, the above description described an example in which the search unit 130 generates a search query itself, searches the corporate information database 150 based on the generated search query, and obtains the corporate information necessary to answer the modified questionnaire. However, the search unit 130 is not limited to this, and may, for example, as described in Embodiment 1, instruct a large-scale language model (LLM) (not shown) to generate a search query and search for corporate information. For example, when the search unit 130 obtains modified questionnaire data from the questionnaire conversion unit 120, it generates a prompt (search prompt) to input into the large-scale language model described above, which includes an instruction to obtain company information necessary to answer the modified questionnaire from the company information database 150. When the search unit 130 generates a search prompt, it inputs the generated search prompt and the modified questionnaire data obtained from the questionnaire conversion unit 120 into the large-scale language model described above. The large-scale language model generates a search query to find company information necessary to answer a modified questionnaire, for example, in accordance with the search prompt. Based on the generated search query, it performs a vector search on the company information database 150 and retrieves the company information necessary to answer the modified questionnaire as a search result. The search unit 130 then outputs the company information retrieved by the large-scale language model as specific company information to the response retrieval unit 140. In this case, the search unit 130 does not perform the search process itself, but rather has the large-scale language model perform the search process, which is expected to reduce the processing load compared to when the search unit performs the search process itself.

[0072] Furthermore, the above description described an example in which the search unit 130 generates a search query itself, searches the corporate information database 150 based on the generated search query, and obtains the corporate information necessary to answer the modified questionnaire, or in which the search unit 130 instructs the large-scale language model to perform these processes. However, even in Embodiment 2, it is not necessarily required that the search unit 130 or the large-scale language model perform the search processing. For example, the information processing device 100 can pre-acquire company information from the company information database 150 that is held by a specified company and is used to answer questions included in a modified questionnaire identified by data acquired by the questionnaire conversion unit 120. Furthermore, if the response acquisition unit 140 can acquire response data using the information acquired in advance, then the search unit 130 or the search process by the large-scale language model becomes unnecessary. In this case, the search procedure (ST130) may be omitted in the information processing program.

[0073] As described above, the information processing program according to Embodiment 2 causes a computer to execute the following: an questionnaire acquisition procedure ST110 for acquiring questionnaire data; a questionnaire conversion procedure ST120 for acquiring data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format, using the questionnaire data acquired in the questionnaire acquisition procedure; and an answer acquisition procedure ST140 for acquiring data indicating answers to questions included in the modified questionnaire, using company information held by a predetermined company for answering questions included in the modified questionnaire identified by the data acquired in the questionnaire conversion procedure ST120, and data indicating the modified questionnaire acquired in the questionnaire conversion procedure ST120. Therefore, the information processing program shown in Embodiment 2 can assist companies in responding to questionnaires created in any format by having them execute predetermined procedures on a computer.

[0074] Furthermore, the information processing program according to Embodiment 2 causes the computer to execute a search procedure ST130 to obtain company information for answering questions included in the modified questionnaire from a company information database 150 that stores company information held by a predetermined company. In the answer acquisition procedure ST140, data indicating answers to the questions included in the modified questionnaire is obtained using the company information obtained in the search procedure ST130 and the data indicating the modified questionnaire obtained in the questionnaire conversion procedure ST120. As a result, the information processing program shown in Embodiment 2 can accurately obtain data indicating answers to questions included in the modified questionnaire by having a computer execute predetermined procedures.

[0075] Furthermore, in the information processing program according to Embodiment 2, the questionnaire conversion procedure ST120 may generate a modification prompt, which is a prompt instructing that the format of the questionnaire identified by the questionnaire data be modified to a predetermined format. The generated modification prompt may be input to the first machine learning model along with the questionnaire data, and data indicating the modified questionnaire may be obtained from the first machine learning model. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute a predetermined procedure and obtain data indicating a modified questionnaire whose format has been appropriately modified to a predetermined format.

[0076] Furthermore, in the information processing program according to Embodiment 2, if the data format of the questionnaire data is not a predetermined data format, before inputting a modification prompt to the first machine learning model, information regarding the arrangement of question fields and answer fields included in the questionnaire, as well as information regarding the content of the questions included in the questionnaire, may be extracted from the questionnaire data, and the extracted information may be included in the modification prompt. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures to obtain data indicating a modified questionnaire in which the format has been appropriately modified to the predetermined format, even if the data format of the questionnaire data is not in the predetermined data format.

[0077] Furthermore, in the information processing program according to Embodiment 2, the questionnaire conversion procedure ST120 may include an instruction in the modification prompt to output data representing the modified questionnaire to the first machine learning model so that the number of question fields and answer fields in the modified questionnaire are equal. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures to obtain data indicating a modified questionnaire in which the number of question fields and answer fields are equal, and can obtain data indicating answers to questions with high accuracy.

[0078] Furthermore, in the information processing program according to Embodiment 2, if the survey conversion procedure ST120 is not in a predetermined data format and the survey data includes image data, the data format of the survey data may be converted to a predetermined data format before inputting the modification prompt to the first machine learning model, and the survey data with the converted data format may be input to the first machine learning model together with the modification prompt. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures to obtain data indicating a modified questionnaire in which the format has been appropriately modified to the predetermined format, even if the data format of the questionnaire data is not in a predetermined data format and the questionnaire data includes image data.

[0079] Furthermore, in the information processing program according to Embodiment 2, the search procedure ST130 may generate a search query based on information indicating questions included in the modified questionnaire identified by the data obtained in the questionnaire conversion procedure, and obtain company information corresponding to the search query from the company information database 150. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures and accurately obtain company information from the company information database 150 for answering questions included in the modified questionnaire.

[0080] Furthermore, in the information processing program according to Embodiment 2, the search procedure ST130 may obtain information indicating a question from the data indicating the modified questionnaire, and generate a search query based on the obtained information indicating the question. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures and generate search queries for searching for company information with high accuracy.

[0081] Furthermore, in the information processing program according to Embodiment 2, the response acquisition procedure ST140 generates a response prompt which is a prompt that includes an instruction to generate responses to questions included in the modified questionnaire identified by the data acquired in the conversion procedure. The company information acquired in the search procedure, the data indicating the modified questionnaire acquired in the questionnaire conversion procedure, and the response prompt are input to a second machine learning model, and data indicating responses to questions included in the modified questionnaire is obtained from the second machine learning model. As a result, the information processing program shown in Embodiment 2 can accurately obtain data indicating answers to questions included in the modified questionnaire by having a computer execute predetermined procedures.

[0082] Furthermore, in the information processing program according to Embodiment 2, the response acquisition procedure ST140 may include an instruction in the response prompt to output the reason for the answer to the question. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures and obtain data indicating the reasons for the answers to questions.

[0083] Furthermore, in the information processing program according to Embodiment 2, if the answer acquisition procedure ST140 is a multiple-choice question, information regarding the options may be included in the answer prompt. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures and reflect information about the choices in the data indicating the answer.

[0084] Furthermore, in the information processing program according to Embodiment 2, the company information obtained in the search procedure ST130 is information disclosed by the company to which the user who answers the questions included in the questionnaire belongs, and the questionnaire data obtained in the questionnaire acquisition procedure ST110 may be obtained from the business partners of the company to which the user belongs. As a result, the information processing program shown in Embodiment 2 can cause a computer to execute predetermined procedures to support the task of responding to questionnaires identified from questionnaire data sent by the user company's business partners.

[0085] Furthermore, in the information processing program according to Embodiment 2, the format of the modified questionnaire may be a format that allows a second machine learning model, which receives data representing the modified questionnaire as input, to process the data and output data representing the answers to the questions included in the modified questionnaire. This allows the information processing program shown in Embodiment 2 to cause a computer to execute predetermined procedures, enabling the output of data indicating the answer by a second machine learning model.

[0086] The information processing device 100 according to Embodiment 1 includes: a questionnaire acquisition unit 110 that acquires questionnaire data; a questionnaire conversion unit 120 that uses the questionnaire data acquired by the questionnaire acquisition unit 110 to acquire data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format; and a response acquisition unit 140 that uses company information held by a predetermined company for answering questions included in the modified questionnaire identified by the data acquired by the questionnaire conversion unit 120, and data indicating the modified questionnaire acquired by the questionnaire conversion unit 120 to acquire data indicating answers to questions included in the modified questionnaire. Therefore, the information processing device 100 shown in Embodiment 1 can assist companies in the process of responding to questionnaires created in any format.

[0087] The information processing method according to Embodiment 1 is an information processing method using an information processing device 100, wherein the questionnaire acquisition unit 110 acquires questionnaire data, the questionnaire conversion unit 120 uses the questionnaire data acquired by the questionnaire acquisition unit 110 to acquire data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format, and the response acquisition unit 140 uses company information held by a predetermined company for answering questions included in the modified questionnaire identified by the data acquired by the questionnaire conversion unit 120, and the data indicating the modified questionnaire acquired by the questionnaire conversion unit 120 to acquire data indicating answers to questions included in the modified questionnaire. By executing the above method, the information processing device 100 can support the response work when a company responds to a questionnaire created in any format. [Explanation of Symbols]

[0088] 100 Information processing device, 110 Questionnaire acquisition unit, 120 Questionnaire conversion unit, 130 Search unit, 140 Response acquisition unit, 150 Corporate information database, 160 First machine learning model, 170 Second machine learning model, 200 Hardware configuration, 210 Communication interface, 220 Input / output interface, 230 Processor, 240 Memory, UT User terminal.

Claims

1. The procedure for obtaining survey data, A survey conversion procedure that uses the survey data obtained in the survey acquisition procedure to obtain data indicating a modified survey in which the format of the survey identified by the survey data has been modified to a predetermined format, A response acquisition procedure that uses company information held by a specified company for answering questions included in a modified questionnaire identified by data obtained in the questionnaire conversion procedure, and data indicating the modified questionnaire obtained in the questionnaire conversion procedure, to acquire data indicating the answers to the questions included in the modified questionnaire. An information processing program that causes a computer to execute something.

2. The computer is instructed to perform a search procedure to retrieve company information from a database that stores company information held by a designated company, in order to answer the questions included in the modified questionnaire. In the aforementioned procedure for obtaining responses, Using the company information obtained in the search procedure and the data representing the modified questionnaire obtained in the questionnaire conversion procedure, data representing the answers to the questions included in the modified questionnaire is obtained. The information processing program according to claim 1.

3. In the aforementioned questionnaire conversion procedure, A modification prompt is generated, which is a prompt instructing that the format of the questionnaire identified by the questionnaire data be modified to the predetermined format. The generated modification prompt is input to a first machine learning model along with the questionnaire data, and data indicating the modified questionnaire is obtained from the first machine learning model. The information processing program according to claim 1 or claim 2.

4. In the aforementioned questionnaire conversion procedure, If the data format of the survey data is not a predetermined data format, before inputting the modification prompt to the first machine learning model, information regarding the arrangement of question fields and answer fields included in the survey, as identified by the survey data, and information regarding the content of the questions included in the survey are extracted from the survey data, and the extracted information is included in the modification prompt. The information processing program according to claim 3.

5. In the aforementioned questionnaire conversion procedure, The modification prompt includes an instruction to cause the first machine learning model to output data representing the modified questionnaire so that the number of question fields and answer fields in the modified questionnaire are equal. The information processing program according to claim 3.

6. In the aforementioned questionnaire conversion procedure, If the data format of the survey data is not a predetermined data format, and the survey data includes image data, the data format of the survey data is converted to the predetermined data format before inputting the modification prompt to the first machine learning model, and the survey data with the converted data format is input to the first machine learning model together with the modification prompt. The information processing program according to claim 3.

7. In the aforementioned search procedure, Based on the information indicating the questions included in the modified questionnaire identified by the data obtained in the aforementioned questionnaire conversion procedure, a search query is generated, and company information corresponding to the search query is obtained from the database. The information processing program according to claim 2.

8. In the aforementioned search procedure, The system obtains information indicating the question from the data representing the modified questionnaire, and generates the search query based on the obtained information indicating the question. The information processing program according to claim 7.

9. In the aforementioned procedure for obtaining responses, In the aforementioned questionnaire conversion procedure, a response prompt is generated, which is a prompt containing instructions to generate answers to questions included in the modified questionnaire identified by the acquired data. The company information for answering the questions included in the modified questionnaire, the data representing the modified questionnaire, and the response prompts are input into a second machine learning model, and data representing the answers to the questions included in the modified questionnaire is obtained from the second machine learning model. The information processing program according to claim 1.

10. In the aforementioned procedure for obtaining responses, The response prompt includes an instruction to output the reason for the answer to the aforementioned question. The information processing program according to claim 9.

11. In the aforementioned procedure for obtaining responses, If the answer format for the aforementioned question is multiple choice, include information about the options in the answer prompt. The information processing program according to claim 9 or claim 10.

12. The company information obtained in the aforementioned search procedure is information disclosed by the company to which the user who answers the questions included in the aforementioned questionnaire belongs. The survey data obtained in the aforementioned survey acquisition procedure is obtained from the client companies of the company to which the user belongs. The information processing program according to claim 2.

13. The format of the aforementioned modified questionnaire is: The second machine learning model, which receives data representing the modified questionnaire as input, processes the data and outputs data representing the answers to the questions included in the modified questionnaire in a format that allows it to output such data. The information processing program according to claim 9.

14. The survey data acquisition unit acquires survey data, A questionnaire conversion unit uses the questionnaire data acquired by the questionnaire acquisition unit to acquire data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format. A response acquisition unit acquires data indicating answers to questions included in a modified questionnaire using company information held by a specified company, which is company information for answering questions included in a modified questionnaire identified by data acquired by the questionnaire conversion unit, and data indicating the modified questionnaire acquired by the questionnaire conversion unit. Information processing device including

15. An information processing method using an information processing device, The survey data acquisition department acquires the survey data. The questionnaire conversion unit uses the questionnaire data acquired by the questionnaire acquisition unit to acquire data indicating a modified questionnaire in which the format of the questionnaire identified by the questionnaire data has been modified to a predetermined format. The response acquisition unit uses company information held by a specified company, which is company information for answering questions included in a modified questionnaire identified by the data acquired by the questionnaire conversion unit, and data indicating the modified questionnaire acquired by the questionnaire conversion unit, to acquire data indicating the answers to the questions included in the modified questionnaire. Information processing methods.

Images