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

Abstract

This invention provides an information processing method, apparatus, and program for evaluating whether a company is taking ESG (Environmental, Social, and Governance) measures into consideration. [Solution] In the information processing method, the program scores evaluation items classified into any of the multiple groups included in the first group group based on publicly available information about the corporation to be evaluated, calculates a first evaluation score for each group based on the scoring results of the evaluation items included in each group of the first group group, scores some of the evaluation items included in the multiple groups that are also classified into any of the multiple groups included in the second group group which is different from the first group group based on publicly available information, calculates a second evaluation score for each group based on the scoring results of the evaluation items included in each group of the second group group, and outputs the calculated first and second evaluation scores in association with the corporation identifier.

Description

Technical Field

[0001] The present invention relates to an information processing method, an information processing apparatus, and an information processing program.

Background Art

[0002] It has been carried out to evaluate whether a company is making efforts considering ESG (Environment, Social, Governance). For example, Patent Document 1 describes an information processing apparatus including an ESG score calculation unit that calculates the ratio of ESG-related words included in a company document based on ESG-related word dictionary data and calculates an ESG score based on the ratio of ESG-related words.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Means for Solving the Problems

[0004] An information processing method according to an aspect of the present invention performs scoring for evaluation items classified into any of a plurality of groups included in a first group group based on public information regarding a corporation to be evaluated, calculates a first evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the first group group, performs scoring for evaluation items that are a part of the evaluation items included in the plurality of groups and are also classified into any of a plurality of groups included in a second group group different from the first group group based on the public information, calculates a second evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the second group group, and a computer executes a process of outputting the calculated first evaluation score and the second evaluation score in association with an identifier of the corporation.

[0005] In the above-described information processing method, the first evaluation score and the second evaluation score are values ​​obtained by converting the total score based on the scoring results into a percentage with a maximum score of 100.

[0006] In the above information processing method, a radar chart based on the first evaluation score for each of the multiple groups included in the first group group, and a radar chart based on the second evaluation score for each of the multiple groups included in the second group group are output.

[0007] In the above information processing method, a radar chart is output based on the first evaluation score and the second evaluation score for multiple companies in the industry to which the corporation belongs.

[0008] In the above-described information processing method, the publicly available information includes data in XBRL format.

[0009] In the above information processing method, the publicly available information in XBRL format and the evaluation items are input into a language model, and the scoring is performed based on the response obtained.

[0010] An information processing device according to one aspect of the present invention is an information processing device equipped with a control unit, the control unit scores evaluation items classified into any of a plurality of groups included in a first group group based on publicly available information relating to a corporation to be evaluated, calculates a first evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the first group group, scores some of the evaluation items included in the plurality of groups that are also classified into any of a plurality of groups included in a second group group different from the first group group based on the publicly available information, calculates a second evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the second group group, and outputs the calculated first evaluation score and second evaluation score in association with the identifier of the corporation.

[0011] An information processing program according to one aspect of the present invention causes a computer to perform the following processes: score evaluation items classified into any of multiple groups included in a first group group based on publicly available information relating to a corporation to be evaluated; calculate a first evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the first group group; score some of the evaluation items included in the multiple groups that are also classified into any of multiple groups included in a second group group different from the first group group based on the publicly available information; calculate a second evaluation score for each group based on the results of the scoring of the evaluation items included in each group of the second group group; and output the calculated first and second evaluation scores in association with the identifier of the corporation.

[0012] It should be noted that the above summary of the invention does not enumerate all of its features. Furthermore, subcombinations of these features may also constitute an invention. [Brief explanation of the drawing]

[0013] [Figure 1] This is an explanatory diagram showing an example configuration of the creation system. [Figure 2] This is a block diagram showing an example of the hardware configuration for Server 1. [Figure 3] This is a block diagram showing the hardware configuration of user terminal 3. [Figure 4] This is an explanatory diagram showing an example of user DB121. [Figure 5] This is an explanatory diagram showing an example of company DB122. [Figure 6] This is an explanatory diagram showing an example of evaluation item DB123. [Figure 7] This is an explanatory diagram showing an example of evaluation item DB123. [Figure 8] This is an explanatory diagram showing an example of evaluation item DB123. [Figure 9] This is an explanatory diagram showing an example of evaluation item DB123. [Figure 10]It is an explanatory diagram showing an example of the evaluation item DB123. [Figure 11] It is an explanatory diagram showing an example of the scoring result DB124. [Figure 12] It is an explanatory diagram showing an example of the evaluation result DB125. [Figure 13] It is an explanatory diagram showing an example of the average DB126. [Figure 14] It is a flowchart showing an example of the business process procedure. [Figure 15] It is a flowchart showing an example of the procedure for setting the target company. [Figure 16] It is a flowchart showing an example of the procedure for information acquisition processing. [Figure 17] It is a flowchart showing an example of the extraction processing procedure. [Figure 18] It is a flowchart showing an example of the scoring processing procedure. [Figure 19] It is an explanatory diagram showing an example of the scoring result screen. [Figure 20] It is an explanatory diagram showing another example of the scoring result screen. [Figure 21] It is an explanatory diagram showing another example of the scoring result screen. [Figure 22] It is an explanatory diagram showing an example of the report display screen. [Figure 23] It is a flowchart showing another example of the information acquisition processing procedure. [Figure 24] It is a flowchart showing another example of the scoring processing procedure. [Figure 25] It is an explanatory diagram showing an example of the scoring prompt. [Figure 26] It is an explanatory diagram showing an example of the comment creation prompt. [Figure 27] It is an explanatory diagram showing an example of the scoring prompt using XBRL data.

Modes for Carrying Out the Invention

[0014] The present invention will be described below through embodiments, but these embodiments are not intended to limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0015] In recent years, ESG (Environmental, Social, and Governance) factors have been used in evaluating companies when making investment decisions. Many companies also publish integrated reports. These reports combine financial and non-financial information. The target audience for integrated reports is broad, including shareholders, investors, business partners, financial institutions, local communities, and employees. The information included in integrated reports also encompasses a company's initiatives from an ESG perspective.

[0016] Traditionally, the information disclosed by companies has primarily been financial information, leading to the belief that collecting non-public information was necessary for a proper evaluation of a company. However, with the increasing disclosure of non-financial information, including integrated reports, company evaluations using publicly available information are also being conducted. This specification proposes a system for creating company evaluation reports using publicly available information. While this specification describes the evaluation target as a company, it is not limited to companies. The evaluation target may also be other legal entities, public interest foundations, general foundations, public interest associations, general associations, specified non-profit organizations, or administrative organizations.

[0017] In this specification, non-financial information (hereinafter also referred to as "non-financial data") is data particularly related to sustainability. Non-financial data is categorized into governance, strategy, risk management, metrics, and targets. Metrics and targets include greenhouse gas emissions for each company. Non-financial data is associated with identifying information that uniquely identifies a company, such as a company ID. Greenhouse gas emissions are, for example, carbon dioxide emissions. Carbon dioxide emissions include direct emissions, indirect emissions, and other indirect emissions. Greenhouse gas emissions may be derived by multiplying at least one usage of electricity, water, oil, and gas by an intensity that indicates the greenhouse gas emissions per unit of each usage.

[0018] As non-financial data, companies may manage information on human rights measures and disaster risk measures related to the products and services they handle. Human rights measures may indicate whether measures are in place to prevent child labor in the manufacturing of products (including not only the assembly of the product itself, but also the assembly of components that make up the product, the processing of materials (including raw materials, etc.) that make up each component, and all other processes related to product manufacturing). Disaster risk measures may indicate whether measures are in place to prevent product manufacturing from being affected in the event of a disaster. In addition to non-financial data, financial data may also be included as collected data, and any other information that the end-user's company wishes to request from the relevant company may be included. For example, various types of damage calculation-based environmental impact assessments, environmental information such as product carbon footprints related to climate change, biodiversity, land use, and raw material procurement related to natural resources, hazardous materials, waste management, and waste such as packaging materials and home appliances related to waste disposal, contained chemical substances, air pollutants, water quality, and soil contamination related to environmentally regulated substances, and other environmental information such as technology, energy, noise, vibration, and odor may be included. Furthermore, information regarding society may include human rights related to human resources (child labor, forced labor, working hours, wages, labor rights, discrimination, etc.), labor management and occupational health and safety, human capital, safety and quality related to the safety of products and services, safety of each substance, privacy and data security, relationships with the community, diversity, equity and inclusion, well-being and engagement, compliance and ethics, and other information related to society. In addition, information regarding governance may include ethics and legal compliance related to corporate conduct, anti-corruption, risk management and disaster response, tax transparency, and other information related to governance. Information regarding due diligence may also be included.

[0019] Non-financial data may include the following information regarding environmental impact: the amounts of by-products, NOx, SOx, BOD, COD, etc. from each of the above-mentioned usage amounts, the impact areas such as air pollution, air pollution, hazardous chemicals, ozone depletion, acidification, noise, global warming, photochemical oxidants, eutrophication, resource consumption, ecotoxicity, fuel consumption, land use, etc., human health based on damage assessment, social projections, biodiversity, primary production, and other protected entities, and the amounts derived using unit consumption factors.

[0020] Direct emissions refer to carbon dioxide emissions (thousand tons of CO2) that fall under Scope 1 of so-called supply chain emissions. Indirect emissions refer to carbon dioxide emissions indirectly emitted by a company through energy purchases during a specified period (fiscal year, quarter, etc.). Indirect emissions refer to carbon dioxide emissions (thousand tons of CO2) that fall under Scope 2 of supply chain emissions. Other indirect emissions refer to carbon dioxide emissions from company activities that are not included in Scope 1 direct emissions or Scope 2 indirect emissions. Other indirect emissions refer to carbon dioxide emissions (thousand tons of CO2) that fall under Scope 3 of supply chain emissions.

[0021] The reduction in carbon dioxide emissions represents the reduction from carbon dioxide emissions during the comparison period. The reduction may be shown separately for direct emissions, indirect emissions, and other indirect emissions. Non-financial data may include the percentage change in emissions (e.g., a 10% decrease compared to the previous year) rather than the reduction from the comparison period.

[0022] (Embodiment 1) Figure 1 is an explanatory diagram showing an example configuration of the creation system. The creation system 100 includes a server 1, an information disclosure site 2, a user terminal 3, and a generation AI service 4.

[0023] Server 1 collects publicly available information about the company to be evaluated from the information disclosure site 2 in response to instructions from user terminal 3, and performs an evaluation of the company. Server 1 transmits the evaluation results to user terminal 3. Server 1 consists of a server computer, workstation, PC (Personal Computer), etc. Server 1 may also be composed of a multicomputer consisting of multiple computers, a virtual machine virtually constructed by software, or a quantum computer. The functions performed by Server 1 may be distributed among multiple computers. Furthermore, the functions of Server 1 may be implemented as a cloud service.

[0024] Information disclosure sites 2 are internet sites that publish a company's financial and non-financial data, such as corporate websites, news sites, or PR (Public Relations) sites. Information disclosure sites 2 also include sites operated by public institutions that publish corporate information for the public good, such as EDINET.

[0025] User terminal 3 is a terminal used by end users. End users are primarily those responsible for evaluating companies, such as consultants and marketing personnel. In Figure 1, two user terminals 3 are shown, but there may be one or three or more.

[0026] Generative AI Service 4 is a service that uses an LLM (Large Language Model) to generate text and images requested by the user. In this specification, it is assumed that Generative AI Service 4 will be used to create evaluation texts from publicly available company information. Note that Generative AI Service 4 is not a required component in this embodiment.

[0027] Figure 2 is a block diagram showing an example of the hardware configuration of Server 1. Server 1 includes a control unit 11, a storage unit 12, a communication unit 13, and a read unit 14. Each component is connected by bus B.

[0028] The control unit 11 has one or more arithmetic processing units such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), and a GPU (Graphics Processing Unit). The control unit 11 reads and executes a program 1P (program product) stored in the storage unit 12, thereby performing various information processing and control processing related to the server 1, and realizing functional units such as the acquisition unit 111, analysis unit 112, scoring unit 113, calculation unit 114, generation unit 115, and output unit 116.

[0029] The acquisition unit 111 acquires publicly available information of the target company from the internet. The analysis unit 112 analyzes the acquired information and extracts the information necessary for evaluation. The scoring unit 113 scores each minimum evaluation item based on the extracted information. The calculation unit 114 calculates the score for each evaluation item. The generation unit 115 creates a report for the target company based on the evaluation results. The output unit 116 displays and outputs the evaluation results and evaluation report.

[0030] The storage unit 12 is composed of SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, or a hard disk or SSD (Solid State Drive). The storage unit 12 temporarily stores data necessary for the control unit 11 to perform calculations. The storage unit 12 also stores the program 1P and various DBs (Databases) necessary for the control unit 11 to perform processing. The storage unit 12 stores the user DB 121, company DB 122, evaluation item DB 123, scoring result DB 124, evaluation result DB 125, and average DB 126. The various DBs stored in the storage unit 12 may be stored on a database server or cloud storage different from server 1.

[0031] The communication unit 13 communicates with the information disclosure site 2, the user terminal 3, and the generation AI service 4 via the network N. Alternatively, the control unit 11 may use the communication unit 13 to download program 1P from another computer via the network N, etc., and store it in the storage unit 12.

[0032] The reading unit 14 reads a portable storage medium 1a, including CD (Compact Disc)-ROM and DVD (Digital Versatile Disc)-ROM. The control unit 11 may read program 1P from the portable storage medium 1a via the reading unit 14 and store it in the storage unit 12. Alternatively, the control unit 11 may read program 1P from a semiconductor memory (not shown).

[0033] Figure 3 is a block diagram showing the hardware configuration of user terminal 3. User terminal 3 consists of a notebook computer, panel computer, tablet computer, smartphone, etc. User terminal 3 includes a control unit 31, a storage unit 32, a communication unit 33, an input unit 34, and a display unit 35. Each component is connected by bus B.

[0034] The control unit 31 has one or more arithmetic processing units such as CPUs, MPUs, and GPUs. The control unit 31 provides various functions by reading and executing programs 3P (programs, program products) stored in the storage unit 32.

[0035] The storage unit 32 is composed of SRAM, DRAM, flash memory, or a hard disk or SSD. The storage unit 32 temporarily stores data necessary for the control unit 31 to perform calculations. The storage unit 32 also stores the program 3P and various databases necessary for the control unit 31 to perform processing. The various databases stored in the storage unit 32 may be stored in a database server or cloud storage.

[0036] The communication unit 33 communicates with the server 1 via the network N. Alternatively, the control unit 31 may use the communication unit 33 to download program 3P from another computer via the network N or the like and store it in the storage unit 32.

[0037] The input unit 34 is a keyboard or mouse. The display unit 35 includes a liquid crystal display panel or an organic EL (electroluminescence) display panel, etc. The display unit 35 displays the evaluation results of the companies being evaluated output by the server 1. Alternatively, the input unit 34 and the display unit 35 may be integrated to form a touch panel display. The user terminal 3 may also display information on an external display device.

[0038] Next, we will describe the database used by the creation system 100. Figure 4 is an explanatory diagram showing an example of the user DB 121. The user DB 121 stores information about end users. The user DB 121 includes a user ID column and a username column. The user ID column stores a user ID that can uniquely identify a user. The user ID may be issued by a computer such as server 1. Alternatively, the user ID may be a string of characters or other characters desired by the end user, as long as it does not overlap with existing users. For example, the email address used by the end user may be used as the user ID. The username column stores the name of the end user. The user DB 121 may also store identification information indicating the organization (corporation or government organization) to which the end user belongs.

[0039] Figure 5 is an explanatory diagram showing an example of Company DB122. Company DB122 primarily stores information about companies to be evaluated. However, it may also store information about companies to which the end users performing the evaluation belong. Company DB122 includes columns for Company ID, Name, Type, Listing Status, Consolidation Status, Securities Code, Corporate Number, Corporate URL, Submitter Code, and Industry. The Company ID column stores a Company ID (identifier) ​​that uniquely identifies a company. The Company ID may be issued by a computer such as Server 1. Alternatively, a string of characters or other characters desired by the company may be used as the Company ID, as long as it does not overlap with existing companies. For example, the domain name of the internet used by the company may be used as the Company ID. The Name column stores the name of the company. The Type column stores the type of company, such as domestic corporation, foreign corporation, etc. The Listing Status column stores whether or not the company is listed on the stock market, if the company is a stock company. The Consolidation Status column stores whether or not there is a parent company or subsidiary that performs consolidated financial statements. The Securities Code column stores the securities code assigned to a company if the company is listed on the stock exchange. The Corporate Number column stores the corporate number assigned to the company. For example, the corporate number is a number assigned by the National Tax Agency. The Corporate URL column stores the URL (Uniform Resource Locator) of the corporate website that the company has made public. The Submitter Code column stores the submitter code assigned by EDINET if the company uses EDINET. The Industry column stores the industry to which the company belongs. The Industry column may also store the classification code of the Japan Standard Industrial Classification. Instead of or in addition to the industry, information on the industry to which the company belongs may be stored in Company DB122. If Company DB122 stores the securities code, corporate number, or submitter code, the securities code, corporate number, or submitter code may be used as the Company ID.

[0040] Figures 6 to 10 are explanatory diagrams showing examples of evaluation item DB123. Evaluation item DB123 stores evaluation items and the points allocated to each item. The evaluation items mainly consist of items related to sustainability, including ESG perspectives. In the creation system 100, the evaluation items have the following characteristics. Major items are labels used to group evaluation items, and evaluation items are essentially defined by medium items, minor items, and minimum items. In this specification, major items are values / long-term strategy, implementation strategy, indicators and targets, governance, and information disclosure. Multiple groups classified by major items are called the first group group. Some evaluation items are grouped by a different perspective (evaluation item 2) than the major items. These groups are classified using evaluation item 2 as a label. In this specification, evaluation item 2 (perspective) is climate change, environment, and society. Multiple groups based on this perspective are called the second group group. That is, all evaluation items belong to one of the groups in the first group group. Some evaluation items belong to both the first group group and the second group.

[0041] The evaluation item DB123 includes columns for Item ID, Major Item, Medium Item, Minor Item, Minimum Item, Score, Evaluation Item 2, and Score 2. The Item ID column stores an Item ID that uniquely identifies each evaluation item. The Major Item, Medium Item, Minor Item, and Minimum Item columns store the content of the evaluation item in four levels. In some cases, the content of an evaluation item can be expressed up to the minor item level. In this case, the Minimum Item will either have no value set, or will store a characteristic value indicating that it is not set, such as blank or "none." The Score column stores the score for the item. In this specification, if the company being evaluated corresponds to the content of an evaluation item, it will be given the score stored in the Score column. If the company being evaluated does not correspond to the content of an evaluation item, the score for that evaluation item will be 0 points. The Evaluation Item 2 column stores classifications from a different perspective than the Major Item (Climate Change, Environment, Society). The Score 2 column stores the score for Evaluation Item 2.

[0042] Figure 6 illustrates some of the records in the evaluation item DB123 where the major category is values / long-term strategy. Evaluation items where the minimum category is climate change are also classified as climate change in evaluation item 2. Evaluation items where the minimum category is water, waste / resources, biodiversity, or chemical management are classified as environment in evaluation item 2. Evaluation items where the minimum category is business, human rights, or human resources are classified as society in evaluation item 2.

[0043] Figure 7 illustrates some records in the evaluation item DB123 where the major item is "implementation strategy." In the examples shown in Figure 7, no records have a value set for the minimum item. Evaluation items where the sub-item is "climate change" are classified as "climate change" in evaluation item 2. Evaluation items where the sub-items are "water," "waste and resources," "biodiversity," and "chemical substance management" are classified as "environment" in evaluation item 2. Evaluation items where the sub-items are "business human rights" and "human resources" are classified as "society" in evaluation item 2.

[0044] Figure 8 illustrates some records in the evaluation item DB123 where the major items are indicators and targets. In the example shown in Figure 8, there are no records where a value is set for the minimum item. An evaluation item where the minor item is "water" is classified as "environment" in evaluation item 2.

[0045] Figure 9 illustrates some of the records in evaluation item DB123 where the major item is governance. In the example shown in Figure 9, there is a mix of records where a value is set in the minimum item and records where it is not. Evaluation items where the sub-item is climate change are also classified as climate change in evaluation item 2.

[0046] Figure 10 illustrates some of the records in the evaluation item DB123 where the major item is "information disclosure." In the example shown in Figure 10, there is a mix of records where a value is set in the minimum item and records where it is not. Also, there are no evaluation items classified under evaluation item 2.

[0047] As described above, all evaluation items are classified into multiple groups within the first group according to the major categories. In addition, some evaluation items are classified into multiple groups within the second group according to evaluation item 2.

[0048] Figure 11 is an explanatory diagram showing an example of the scoring results DB124. The scoring results DB124 stores the results of scoring the evaluated company according to the evaluation items. The scoring results DB124 includes a company ID column, an item ID column, a response column, a major item column, a score column, an evaluation item 2 column, and a score 2 column. The company ID column stores the company ID of the company being evaluated. The item ID column stores the item ID of the evaluation item. The response column stores the response to the evaluation item. Here, the response column stores 1 if the evaluated company fits the content of the evaluation item, and 0 if it does not. The major item column stores the major item corresponding to the evaluation item. The score column stores the score for the evaluation item. If the value in the response column is 1, the score column stores the score stored in the scoring column of the evaluation item DB123 as the score. If the value in the response column is 0, the score column stores 0 as the score. The evaluation item 2 column stores the classification of the second group. The Score 2 column stores the score for evaluation item 2. If the value in the Answer column is 1, the Score column stores the score stored in the Score 2 column of evaluation item DB123. If the value in the Answer column is 0, the Score column stores 0 as the score.

[0049] Figure 12 is an explanatory diagram showing an example of the evaluation results DB125. The evaluation results DB125 stores evaluation values ​​for each major category and each evaluation item 2 based on the scoring results. The evaluation results DB125 includes a company ID column, an item column, and an aggregation column. The company ID column stores the company ID of the company being evaluated. The item column stores the major category or evaluation item 2. The aggregation column stores the aggregation result and evaluation score for the major category or evaluation item 2. Since the evaluation items included in the major category and evaluation item 2 are not constant, if the sum of the scores (total score) is used as the evaluation score, comparisons between major categories and between evaluation items 2 are not possible. Therefore, the evaluation score is expressed as a percentage with a maximum score of 100.

[0050] Figure 13 is an explanatory diagram showing an example of Average DB126. Average DB126 stores the average of the evaluation values ​​for each industry that serve as comparison data. Details will be described later. Average DB126 includes a classification code column, an industry column, an item column, and an average value column. The classification code column stores the classification code indicating the industry. For example, the classification code used is from the Japan Standard Industrial Classification. 319 shown in Figure 13 is the code for the sub-classification indicating the software industry. The industry column stores the industry. The item column stores the evaluation item or evaluation item 2. The average value column stores the average value for each item. In addition to the average value, the maximum value, minimum value, mode, median, etc., may also be calculated and stored.

[0051] Next, we will explain the information processing performed in the creation system 100. Figure 14 is a flowchart showing an example of the business processing procedure. The end user operates the user terminal 3 to input or select the identification information of the company to be evaluated. The identification information includes the company name, the company's securities code, etc. If the creation system 100 has assigned a company ID and the end user is aware of it, they may input that company ID. The control unit 31 of the user terminal 3 sends the identification information to the server 1. The control unit 11 of the server 1 receives the identification information. The control unit 11 sets the company to be evaluated (step S1). The control unit 11 acquires information about the company to be evaluated (step S2). The control unit 11 extracts the information required for each evaluation item from the acquired information (step S3). The control unit 11 scores the evaluation items and the extracted information (step S4). The control unit 11 calculates the evaluation score for each major item (first evaluation score) and the evaluation score for each evaluation item 2 (second evaluation score) from the scoring results (step S5). The control unit 11 outputs the evaluation results to the user terminal 3 (step S6). It is desirable that the evaluation results include not only the evaluation score, but also the scoring results for each evaluation item and the information that formed the basis of the scoring. The end user checks the evaluation results displayed on the user terminal 3 and decides whether correction is necessary. If the end user decides that correction is necessary, they input a correction instruction to the user terminal 3. If the end user decides that correction is not necessary, they input an approval of the evaluation results to the user terminal 3. The user terminal 3 sends the decision result, either a correction instruction or approval, to the server 1. The control unit 11 of the server 1 receives the decision result and decides whether correction is necessary (step S6). If the control unit 11 decides that correction is necessary (YES in step S7), it makes corrections according to the correction instructions (step S8). The correction here is assumed to be the scoring results for each evaluation item. The control unit 11 returns to step S5. If the control unit 11 decides that correction is not necessary (NO in step S7), it stores the evaluation results (step S9) and terminates the process.

[0052] Figure 15 is a flowchart showing an example of the procedure for setting up target companies. The process for setting up target companies corresponds to step S1 in Figure 14. The control unit 11 of server 1 searches the company database 122 using the identification information of the company to be evaluated received from the user terminal 3 and determines whether or not the company is registered (step S21). If the control unit 11 determines that the company is not registered (NO in step S21), it obtains basic information about the company to be evaluated that should be stored in the company database 122 (step S22).

[0053] Basic information can be obtained in the following ways, for example: Send an input screen to user terminal 3 for the end user to input the information. The end user inputs the basic information into user terminal 3 and issues a send command. User terminal 3 sends the entered basic information to server 1. Server 1's control unit 11 receives the basic information. Another method is to use an internet search engine. Server 1's control unit 11 creates a search query that includes the identification information of the company to be evaluated, received from user terminal 3, and obtains the basic information from the results obtained by submitting the created search query to the search engine. Information that cannot be obtained by the search engine can be entered by the end user using the method described above.

[0054] The control unit 11 stores the acquired basic information in the company database 122 (step S23). The control unit 11 temporarily stores the basic information for use in later processing (step S24) and then terminates.

[0055] If the control unit 11 determines that the information is registered (YES in step S21), it reads the basic information from the company DB 122 (step S25). The control unit 11 temporarily stores the read basic information for use in later processing (step S24) and then terminates.

[0056] Figure 16 is a flowchart showing an example of the information acquisition process procedure. The information acquisition process corresponds to step S2 in Figure 14. The control unit 11 of server 1 acquires information about the company to be evaluated using the URL of the company's corporate website, which is temporarily stored (step S31). The information to be acquired includes, for example, ESG reports, sustainability reports, CSR (Corporate Social Responsibility) reports, and integrated reports. Information acquisition can be achieved using publicly known technologies such as crawling and scraping. The control unit 11 acquires securities reports, quarterly reports, etc. from EDINET using the submitter code of the company to be evaluated, which is temporarily stored (step S32). The control unit 11 may also acquire the information using an API (Application Programmable Interface). The control unit 11 collects information about the company to be evaluated from PR (Public Relations) sites and news sites. It is desirable to set the sites from which to collect information in advance. Trustworthy sites should be set to prevent the acquisition of false information such as fake news. The control unit 11 stores the acquired information in the storage unit 12, etc. (step S34) and terminates the process. Furthermore, if the acquired information includes image data such as graphs, character recognition (CSI) is performed to convert the included information into text and store it. Similarly, if a document is acquired as image data, character recognition is also performed.

[0057] Figure 17 is a flowchart showing an example of the extraction process procedure. The extraction process corresponds to step S3 in Figure 14. The control unit 11 of server 1 selects the evaluation items to be processed (step S41). The control unit 11 extracts company information that matches the evaluation items from the entire amount of company information already collected (step S42). For example, the evaluation items and company information are vectorized, and company information whose cosine similarity with the evaluation items is above a threshold is extracted. The control unit 11 associates the extracted company information with the evaluation items and stores it in a temporary storage area provided in the storage unit 12, etc. (step S43). The control unit 11 determines whether or not there are any unprocessed evaluation items (step S44). If the control unit 11 determines that there are unprocessed evaluation items (YES in step S44), it returns the process to step S41 and processes the unprocessed evaluation items. If the control unit 11 determines that there are no unprocessed evaluation items (NO in step S44), it returns the process to the caller.

[0058] Figure 18 is a flowchart showing an example of the scoring process procedure. The scoring process corresponds to step S4 in Figure 14. The control unit 11 of server 1 obtains the content of the evaluation items to be processed (step S51). The control unit 11 obtains company information corresponding to the evaluation items from the temporary storage area (step S52). The control unit 11 makes a determination on the evaluation items from the obtained company information (step S53). The control unit 11 stores the determination result in the scoring result DB 124 (step S54). The control unit 11 determines whether or not there are any unprocessed evaluation items (step S55). If the control unit 11 determines that there are unprocessed evaluation items (YES in step S55), it returns the process to step S51 and processes the unprocessed evaluation items. If the control unit 11 determines that there are no unprocessed evaluation items (NO in step S55), it returns the process to the caller.

[0059] Figure 19 is an explanatory diagram showing an example of the scoring results screen. The scoring results screen d01 includes a list table d011 and a confirmation button d012. The list table d011 displays the scoring results in a list. The list table d011 includes columns for major, medium, minor, minimum, scoring, answer, score, evaluation item 2, scoring 2, score 2, and data. The major, medium, minor, and minimum columns represent the major, medium, minor, and minimum items of the evaluation items, respectively. The scoring column shows the scoring for the evaluation items. The answer column shows the scoring results for the evaluation items of the first group. If the company being evaluated corresponds to an evaluation item, it is indicated with a circle (〇). If the company being evaluated does not correspond to an evaluation item, it is indicated with a minus sign (-). The score column shows the score for each evaluation item. If the answer column is a circle, the value from the scoring column is shown. The evaluation item 2 column shows the evaluation items of the second group (hereinafter also referred to as "evaluation item 2"). If not applicable, the field will be left blank. Column 2, "Score Distribution," shows the points allocated to evaluation item 2. Column 2, "Score," shows the score for each evaluation item 2. If the "Answer" column is circled, the value from Column 2 is displayed. Column 2, "Materials," shows hyperlinks to referenced materials (company information) used for scoring each evaluation item. If there are multiple materials, multiple hyperlinks may be shown in the "Materials" column. Pressing the "Confirm" button (d012) finalizes the displayed scoring results, making them unchangeable.

[0060] Figure 20 is an explanatory diagram showing another example of the scoring results screen. Figure 20 shows the screen when an end user changes the scoring results. Components similar to those in Figure 19 are given the same symbols and explanations are omitted. In Figure 20, the end user is attempting to change the answers to the evaluation items, which are the major category "Values" and the subcategory "Identification of Materiality," from circles to minus (-) using the pull-down menu d013. Since these evaluation items do not fall under evaluation item 2, the score for 2 is 0 even though the answer is a circle.

[0061] Figure 21 is an explanatory diagram showing another example of the scoring results screen. Figure 21 shows the screen after the end user has changed the scoring results, but before the changes are reflected. Components similar to those in Figure 19 are given the same symbols and their explanations are omitted. As described above, in Figure 21, the answers to the evaluation items, which are the major category "Values" and the subcategory "Identification of Materiality," have been changed to negative (-). Accordingly, the score for these items has changed from 10 to 0. Also, the confirmation button d012 has changed to the edit button d014. When the edit button d014 is operated, the changes are sent from the user terminal 3 to the server 1, and the changes are reflected.

[0062] Up to this point, we have explained how end-users define the companies to be evaluated and conduct the evaluation. In order to accurately understand the evaluation results of the companies to be evaluated, comparative data is necessary. The following explains comparative data. Since the evaluation by the creation system 100 is based on publicly available information, if the evaluation results are understood as absolute evaluations, they may lack accuracy and appropriateness. Therefore, by comparing them with comparative data, a certain degree of accuracy and appropriateness can be ensured.

[0063] The comparative data will be, for example, the average of the evaluation results of multiple companies. The reason for using evaluation results from multiple companies is to ensure, to some extent, the validity of the comparative data. The evaluation method is the same as the method used for the company being evaluated.

[0064] Comparative data will be created for each industry. This is because various circumstances are expected to differ depending on the industry, and it is believed that comparisons within the same industry will allow for more accurate judgments. The selection of multiple companies (multiple companies) for creating comparative data will be, for example, the top 10 companies in terms of sales. Selection may also be based on factors other than sales, such as net profit, number of employees, or total assets. The selected multiple companies will be evaluated, and the evaluation results and their average values ​​will be stored in the average DB126 of the storage unit 12 of server 1.

[0065] Figure 22 is an explanatory diagram showing an example of a report display screen. The report display screen d02 includes company information d021, an overall evaluation chart d022, a data collection status chart d023, an evaluation results list d024, industry characteristics d025, and proposed content d026. Company information d021 indicates the company being evaluated. Company information d021 includes at least the company name. Figure 22 also shows the industry and CDP score. The CDP score is the score published by CDP (Carbon Disclosure Project). The overall evaluation chart d022 shows the evaluation results based on the evaluation items of the first group using a radar chart. The data collection status chart d023 shows the evaluation results based on the evaluation items of the second group (evaluation item 2) using a radar chart. The evaluation results list d024 shows comments on the evaluation results. Industry characteristics d025 shows industry-specific circumstances. If there are no particular circumstances, general information may be shown. Proposed content d026 shows the services that the company to which the end user conducting the evaluation belongs can provide.

[0066] This embodiment offers the following advantages: It makes it possible to evaluate target companies from a sustainability perspective using publicly available information (primarily non-financial information). Furthermore, by presenting comparative data, it becomes possible to accurately grasp the evaluation results.

[0067] (modified version) This modified example describes the use of XBRL format data. For example, if the company being evaluated issues an integrated report, it is evaluated whether the issue date is the same as the securities report, within three months of the securities report, or more than three months after the securities report. When the securities report is obtained from EDINET in XBRL format, the control unit 11 of server 1 can obtain the issue date of the securities report (more precisely, the submission date to EDINET) by searching for the element with the element name jpcrp_cor:FilingDateCoverPage and context ID FilingDateInstant. Furthermore, if the integrated report is also available in XBRL format, the issue date can be obtained in a similar manner. If the issue dates of the integrated report and the securities report can be obtained, the above determination can be made.

[0068] Furthermore, when evaluating whether Scope 1 and 2 disclosures of GHG (Green House Gas) emissions are made in a securities report, it is possible to determine whether Scope 1 emissions are disclosed by searching for the element name jpcrp_cor:GrossScope1GreenhouseGasEmissionsScope1And2GreenhouseGasEmissions. Similarly, it is possible to determine whether Scope 2 emissions are disclosed by searching for the element name jpcrp_cor:GrossScope2GreenhouseGasEmissionsScope1And2GreenhouseGasEmissions.

[0069] As described above, by using XBRL format data, it becomes possible to have the control unit 11 of server 1 perform scoring using a relatively simple logic, depending on the content of the evaluation items. If the non-financial information necessary for the evaluation can be obtained as XBRL format data, it becomes easy to extract data for each evaluation item.

[0070] When using XBRL format data, the system is designed to extract the necessary data for each evaluation item using logic, but this is not the only option. The generation AI service 4 may also be used to extract and score the data. The control unit 11 of server 1 creates a prompt that includes the evaluation item, the XBRL format data, and the name of the element to be referenced. The control unit 11 sends the prompt to the generation AI service 4 and receives the scoring result.

[0071] (Embodiment 2) This embodiment relates to evaluation using the generation AI service 4. The following description will mainly explain the differences between this embodiment and Embodiment 1. First, the business processing and target company setting processing are the same as in Embodiment 1.

[0072] The information acquisition process in this embodiment will now be described. Figure 23 is a flowchart showing another example of the information acquisition process. Steps that are common with the content shown in Figure 16 are given the same step number and their explanation is omitted. The control unit 11 of server 1 performs steps S31 to S33. The control unit 11 sends the image contained in the acquired company information to the generation AI service 4 and performs image recognition (step S35). The generation AI service 4 creates a descriptive text of the image content, and if the image is a graph, it generates text of the content that the graph represents, such as axes and values. The control unit 11 receives the descriptive text and graph content test from the generation AI service 4. The control unit 11 indexes the company information (step S36). For each set of data that constitutes the company information, the control unit 11 generates an index that includes search keywords and summaries. The control unit 11 may also have the generation AI service 4 perform the process of dividing the company information into sets. The control unit 11 vectorizes (embedding) the company information (step S37). The control unit 11 associates the vectorized company information with the generated index and stores it in the storage unit 12 (step S34).

[0073] Figure 24 is a flowchart showing another example of the scoring process procedure. The control unit 11 of server 1 acquires the evaluation items to be processed (step S71). The control unit 11 searches for company information (step S72). The control unit 11 performs an index search using words included in the evaluation items as search terms. The control unit 11 vectorizes the content of the evaluation items and searches for information similar to the content of the evaluation items from the vectorized company information. For similarity determination, for example, cosine similarity is used. The content of the evaluation items may be vectorized in advance and stored in the storage unit 12. The control unit 11 creates a prompt (step S73). The prompt includes a question about the evaluation items and the company information obtained as a result of the search in step S72. The control unit 11 sends the created prompt to the generation AI service 4 (step S74). The control unit 11 receives a response from the generation AI service 4 (step S75). The control unit 11 stores the response and score in the scoring result DB 124 (step S76). If evaluation item 2 is also applicable, the control unit 11 also stores score 2 in the scoring result DB 124. The control unit 11 determines whether or not there are any unprocessed evaluation items (step S77). If the control unit 11 determines that there are unprocessed evaluation items (YES in step S77), it returns to step S71 and processes the unprocessed evaluation items. If the control unit 11 determines that there are no unprocessed evaluation items (NO in step S77), it returns the process to the caller.

[0074] The prompt created in step S73 may be as follows. Figure 25 is an explanatory diagram showing an example of a prompt. Figure 25A shows an example of a prompt that includes questions about evaluation items and company information. Figure 25B shows an example of a prompt that, in addition to the scoring result, is answered by the generating AI service 4.

[0075] The generation AI service 4 may be used to create the report shown in Figure 22. For example, the overall evaluation chart d022 and the data collection status chart d023 may be created by the generation AI service 4. The evaluation results shown in the evaluation results list d024 may also be generated by the generation AI service 4. For industry characteristics d025, for example, the evaluation results of multiple companies evaluated to create comparative data may be referenced by the generation AI service 4 to generate them. Furthermore, the generation AI service 4 may be used to generate proposal content d026 from the overall evaluation chart d022, the data collection status chart d023, the evaluation results list d024, the industry characteristics d025, and the services that the company to which the end user belongs can provide.

[0076] Next, we will explain the prompts used to give instructions to the generation AI service 4. Figure 25 is an explanatory diagram showing an example of a scoring prompt. Figure 25A is an example of a prompt that returns only the scoring result. As shown in Figure 25A, by sending a prompt that only asks for scoring to the generation AI service 4, it becomes possible to have the generation AI service 4 perform the scoring. The [Company Information] section contains vectorized information of the companies to be evaluated, as well as URIs (Uniform Resource Identifiers) that indicate the location of the information of the companies to be evaluated.

[0077] Figure 25B shows an example of a prompt that returns the scoring result and its rationale. As shown in Figure 25B, the rationale for the result is returned from the generating AI service 4, allowing the end user to verify whether there are any errors in the scoring result based on the scoring result and its rationale.

[0078] Figure 26 is an explanatory diagram showing an example of a comment creation prompt. This prompt is used to have the generation AI service 4 create the comments to be described in industry characteristic d025 shown in Figure 22. [Industry Information] should contain vectorized information of the collected information on the companies being evaluated, a URI indicating the location of the information on the companies being evaluated, or a URI of a source of information to which information on the industry to which the companies being evaluated belong can be obtained. Sources of information include industry publications, the URL of the corporate website of the company from which the comparative data was created (Uniform Resource Locator), etc. URLs of news sites that disseminate information on sustainability trends are also acceptable.

[0079] Figure 27 is an explanatory diagram showing an example of a scoring prompt using XBRL data. As shown in Figure 27, even when the generation AI service 4 is made to handle data in XBRL format, it is believed that highly accurate scoring results can be obtained by indicating the element names of the data to be referenced.

[0080] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention.

[0081] It should be noted that the execution order of operations, procedures, steps, and stages in the apparatus, systems, programs, and methods shown in the claims, specifications, and drawings is not explicitly stated as "before," "prior to," etc., and that these can be implemented in any order unless the output of a previous process is used in a later process. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," "next," etc. for convenience, it does not mean that it is essential to perform the operations in that order.

[0082] Furthermore, aspects of each embodiment can be embodied in whole or in part by a computer. For example, a program installed on such a computer may cause the computer to function as an operation associated with an apparatus according to an embodiment of the present invention, or as one or more "parts" of such apparatus. Alternatively, the program may cause the computer to execute such operation or one or more "parts." The program may cause the computer to execute a process or a stage of such process according to an embodiment of the present invention. Such a program may be executed by the CPU to cause the computer to execute a particular operation associated with some or all of the blocks in the flowcharts and block diagrams described herein. [Explanation of Symbols]

[0083] 100: Creation System 1: Server 11: Control Unit 111: Acquisition Department 112:Analysis Department 113: Grading Department 114: Calculation Unit 115: Generation part 116: Output section 12: Storage section 121: User DB 122: Corporate DB 123: Evaluation Item Database 124: Scoring Results Database 125: Evaluation Results DB 126: Average DB 13: Communications Department 14: Reading section 1P: Program 1a: Portable storage medium 2: Information disclosure site 3: User terminal 31: Control Unit 32: Storage section 33: Communications Department 34: Input section 35: Display section 3P: Program 4: Generative AI service B: Bus N: Network

Claims

1. Based on publicly available information concerning the corporations to be evaluated, scores are assigned to evaluation items classified into one of the multiple groups included in Group 1. Based on the scoring results of the evaluation items included in each of the first group groups, a first evaluation score is calculated for each group. Based on the aforementioned publicly available information, a portion of the evaluation items included in the aforementioned multiple groups, which are classified into any of the multiple groups included in the second group, which is different from the first group, will be scored. Based on the scoring results of the evaluation items included in each of the second group groups, a second evaluation score is calculated for each group. The calculated first evaluation score and second evaluation score are output in association with the identifier of the corporation. An information processing method in which a computer performs the processing.

2. The first and second evaluation scores are calculated by converting the total score based on the scoring results into a percentage with a maximum score of 100. The information processing method according to claim 1.

3. The system outputs radar charts based on the first evaluation scores for each of the multiple groups included in the first group group, and radar charts based on the second evaluation scores for each of the multiple groups included in the second group group. The information processing method according to claim 1 or claim 2.

4. Output a radar chart based on the first and second evaluation scores for multiple companies in the industry to which the aforementioned corporation belongs. The information processing method according to claim 3.

5. The aforementioned publicly available information includes data in XBRL format. The information processing method according to claim 1 or claim 2.

6. The aforementioned publicly available information in XBRL format and the aforementioned evaluation items are input into a language model, and the scoring is performed based on the response obtained. The information processing method according to claim 5.

7. An information processing device equipped with a control unit, The control unit, Based on publicly available information concerning the corporations to be evaluated, scores are assigned to evaluation items classified into one of the multiple groups included in Group 1. Based on the scoring results of the evaluation items included in each of the first group groups, a first evaluation score is calculated for each group. Based on the aforementioned publicly available information, a portion of the evaluation items included in the aforementioned multiple groups, which are classified into any of the multiple groups included in the second group, which is different from the first group, will be scored. Based on the scoring results of the evaluation items included in each of the second group groups, a second evaluation score is calculated for each group. The calculated first evaluation score and second evaluation score are output in association with the identifier of the corporation. An information processing device that performs processing.

8. Based on publicly available information concerning the corporations to be evaluated, scores are assigned to evaluation items classified into one of the multiple groups included in Group 1. Based on the scoring results of the evaluation items included in each of the first group groups, a first evaluation score is calculated for each group. Based on the aforementioned publicly available information, a portion of the evaluation items included in the aforementioned multiple groups, which are classified into any of the multiple groups included in the second group, which is different from the first group, will be scored. Based on the scoring results of the evaluation items included in each of the second group groups, a second evaluation score is calculated for each group. The calculated first evaluation score and second evaluation score are output in association with the identifier of the corporation. An information processing program that instructs a computer to perform a task.

Images