Information processing device, information processing system, and information processing method

Abstract

The aim is to facilitate the disclosure of ESG information. [Solution] The information processing device 10 takes ESG (environmental, social, and governance) information as input and converts the ESG information according to predetermined standards. Using a pre-prepared learning model, it converts the input ESG information to conform to predetermined standards and creates a prompt. Based on this prompt, an ETL (extract, transform, write) processing script generates ESG information that conforms to the predetermined standards.

Description

Technical Field

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

Background Art

[0002] For example, in ESG (Environmental, Social, Governance) rating, a system has been proposed that objectively and appropriately evaluates corporate value by quantifying non-financial information (Patent Document 1).

[0003] In addition, in the introduction of ESG management, various disclosure systems that require disclosure in accordance with disclosure standards have been established. In ESG information disclosure, the burden on the ESG personnel of enterprises is large, and technologies to support this are desired.

[0004] Today, systems that support ESG management by enterprises have also emerged. For example, there is a proposal for a system that quantitatively collects ESG information of a specific enterprise as data and outputs information based on the data (for example, the achievement degree of ESG goals of each index, each of the ESG data and financial data of each index, and the result of correlation analysis thereof, future prediction) (Patent Document 2).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] Even with the conventional technologies described above, ESG information disclosure requires the collected ESG information to be converted into an international standard format (for example, SASB: Sustainability Accounting Standards Board). Therefore, users must manually map the relationships between the different formats, which presents a challenge due to the amount of effort required. Furthermore, regarding methods of disclosing ESG information, 2024 is a transitional period, with multiple standards existing in addition to international standards. Companies and supporting businesses need to disclose ESG information according to their respective standards, and this process can be burdensome.

[0007] Furthermore, regarding ESG information disclosure standards such as international standards, there is a challenge in that it becomes difficult to compare the latest version of ESG information with older versions when the version changes.

[0008] This invention was made in view of the above background, and aims to provide an information processing device, an information processing system, and an information processing method that can facilitate the disclosure of ESG information. [Means for solving the problem]

[0009] To solve the above-mentioned problems and achieve the above objectives, one embodiment of the present invention is an information processing device comprising: a storage unit that stores a mapping model, vector data, and script model that have learned the correspondence between ESG information and predetermined criteria; a mapping creation unit that references the mapping model and vector data and creates mapping data based on input ESG information; a prompt creation unit that creates prompts based on the mapping data created by the mapping creation unit; a generation unit that references a script model and generates an ETL (extract, transform, write) processing script based on the prompts created by the prompt creation unit; and a conversion unit that converts the ETL processing script generated by the generation unit into ESG format to generate ESG information.

[0010] Another embodiment of the present invention is an information processing system comprising: an input device for inputting ESG information; an information processing device for converting the ESG information input by the input device according to predetermined criteria; and an output device for outputting the ESG information converted by the information processing device, wherein the information processing device uses a pre-prepared learning model to convert the ESG information input by the input device according to predetermined criteria to create a prompt, and generates ESG information according to the predetermined criteria using an ETL processing script based on the prompt.

[0011] Furthermore, another embodiment of the present invention is an information processing method for performing information processing on a computer, characterized by comprising: a mapping creation step of creating mapping data based on input ESG information by referring to a mapping model and vector data stored in memory that have learned the correspondence between ESG information and predetermined criteria; a prompt creation step of creating prompts based on the mapping data created in the mapping creation step; a generation step of generating an ETL processing script based on the prompts created in the prompt creation step by referring to a script model stored in memory; and a conversion step of converting the ETL processing script generated in the generation step into ESG format to generate ESG information. [Effects of the Invention]

[0012] According to the present invention, the disclosure of ESG information becomes easier. [Brief explanation of the drawing]

[0013] [Figure 1] This is a block diagram showing an example of the functionality of an information processing device according to one embodiment of the present invention. [Figure 2] This is a configuration diagram showing an example of the hardware configuration of an information processing device according to one embodiment of the present invention. [Figure 3] This is an explanatory diagram illustrating an example of domain knowledge using one embodiment of the present invention. [Figure 4] This is an explanatory diagram illustrating the main sequence according to one embodiment of the present invention. [Figure 5] This is an explanatory diagram for explaining a processing procedure according to an embodiment of the present invention. [Figure 6] This is an explanatory diagram for explaining a processing procedure according to an embodiment of the present invention. [Figure 7] This is an explanatory diagram for explaining a processing procedure according to an embodiment of the present invention. [Figure 8] This is an explanatory diagram for explaining a specific example of processing according to an embodiment of the present invention. [Figure 9] This is an explanatory diagram for explaining a specific example of processing according to an embodiment of the present invention. [Figure 10] This is an explanatory diagram for explaining a specific example of processing according to an embodiment of the present invention. [Figure 11] This is an explanatory diagram for explaining a specific example of processing according to an embodiment of the present invention. [Figure 12] This is an explanatory diagram for explaining a specific example of processing according to an embodiment of the present invention.

Mode for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the following description and drawings are merely examples for explaining the present invention, and for the sake of clarity of explanation, appropriate omissions and simplifications have been made. Also, the present invention can be implemented in various other forms. Further, unless otherwise limited, each component may be singular or plural.

[0015] In the following description, the same or similar configurations may be denoted by the same reference numerals and redundant descriptions may be omitted. Also, in the following description, various types of information may be described using expressions such as "information" and "table", but the various types of information may be represented by other data structures. Also, as expressions for identification information, there are expressions such as "identification information", "identifier", "name", "ID", and "number", and these can be replaced with each other. Also, in the following description, "database" is abbreviated as "DB" and "table" is abbreviated as "TBL".

[0016] First, the configuration will be described using FIGS. 1 and 2. FIG. 1 is a block diagram showing a functional example of the information processing apparatus according to the present embodiment, and FIG. 2 is a configuration diagram showing a hardware configuration example of the information processing apparatus according to the present embodiment.

[0017] As shown in FIG. 2, for example, the information processing apparatus 10 is connected via a network (not shown) to an ETL (Extract, Transform, Load) operator terminal 20, an ESG person-in-charge terminal 30 operated by an ESG information disclosure person-in-charge, a technician terminal 50, etc., as examples of external devices, and performs processing for disclosing ESG information in the latest international standard format.

[0018] The ETL operator terminal 20, the ESG person-in-charge terminal 30 operated by the ESG information disclosure person-in-charge, and the technician terminal 50 all include a keyboard for operation and data input in the present embodiment and a display for screen display in the present embodiment. Of course, a tablet type integrating the keyboard and the display may also be used.

[0019] The technician terminal 50 is connected to a domain knowledge DB 60 and an international standard DB 70 via a network (not shown) and provides domain knowledge data and international standard data to be stored in a vector DB 122 described later.

[0020] Note that the information processing apparatus 10, the ETL operator terminal 20, and the ESG person-in-charge terminal 30 form an information processing system, and converted ESG information is generated from the ESG information before conversion within this information processing system.

[0021] The information processing apparatus 10 is composed of a communication unit 11, a storage unit 12, a mapping creation unit 13, a prompt creation unit 14, an ETL process generation unit 15, an ETL process script verification unit 16, an ESG format conversion unit 17, and the like.

[0022] The communication unit 11 controls data communication between the inside of the information processing apparatus 10 and the above-described external devices. The storage unit 12 is a storage that stores a mapping model 121, a vector DB 122, a script model 123, and the like.

[0023] Mapping Model 121 functions as an LLM (Large-Scale Language Model) and is a generative AI model used to create mapping data using unit conversion rules between source and destination data items for ESG information. Vector DB 122 functions as a RAG (Retrieval-Augmented Generation) and is a database used to convert mapping data into vector data. Script Model 123 functions as a generative AI (Artificial Intelligence) and is a generative AI model used to generate ETL processing script data in response to prompts.

[0024] The mapping creation unit 13 uses the mapping model 121 to create mapping data based on unit conversion rules between source data items and converted data items for ESG information, and then uses the vector DB 122 to convert that mapping data into script data.

[0025] The prompt generation unit 14 generates prompts to ask the script model 123 questions based on the script data created by the mapping generation unit 13, and the ETL processing generation unit 15 obtains ETL processing script data from the script model 123 as an answer based on the prompts.

[0026] The ETL processing script verification unit 16 verifies the ETL processing script data, and the ESG format conversion unit 17 converts the ETL processing script data to ESG format based on the verification results of the ETL processing script verification unit 16, and outputs the converted ESG information to the ESG person's terminal 30 via the communication unit 11. The storage unit 12 may store the converted ESG information before and after the conversion in association with each other.

[0027] The information processing device 10 is composed of, for example, a CPU 101 that controls the entire device, a memory 103 that stores programs 102 such as processes to be executed by the CPU 101 (see Figure 2) and various data being executed, an operating device 104 equipped with a keyboard and a display device, an external storage device 105 that registers and stores various data in formats such as DB and TBL, a communication IF 106 connected to the network 40 and responsible for communication with various external devices, and a bus 107 connected to each unit within the device and responsible for communication of data, signals, etc. within the device.

[0028] Here, we will explain domain knowledge using Figure 3. Figure 3 is an explanatory diagram illustrating an example of domain knowledge according to this embodiment. As shown in Figure 3, the domain knowledge DB60 is configured to store items indicating the measurement target in ESG information and their measurement units, corresponding to codes.

[0029] For example, the code "TC001" is associated with the item "Total Energy Consumption" and its unit of measurement is "Joule". The code "TC002" is associated with the item "Total Water Intake" and its unit of measurement is "Cubic Meter".

[0030] Additionally, codes "TC003" and "TC004" are associated with the items "Considerations on integrating environmental considerations into strategic planning for data center needs" and "Explanation of policies and practices regarding behavior-based advertising and user privacy," respectively, with the unit of measurement being "n / a" (not applicable).

[0031] Furthermore, the code "TC005" is associated with the item "Number of users who use the information for secondary purposes," and its unit of measurement is "Number of cases." The code "TC006" is associated with the item "Total amount of financial loss resulting from legal proceedings concerning user privacy," and its unit of measurement is "Reported currency."

[0032] Furthermore, the code "TC007" is associated with the following items: "Number of requests from law enforcement agencies for user information, number of users for whom information was requested, and percentage of cases where disclosure was granted," with "percentage" as the unit of measurement.

[0033] Next, the operation will be explained using Figure 4. Figure 4 is an explanatory diagram illustrating the main sequence according to this embodiment. The sequence shown in Figure 4 shows an example of processing that takes place between the ETL worker terminal 20 operated by the ETL creator, the ESG person terminal 30, the engineer terminal 50, the information processing device 10, the mapping model 121, the vector DB 122, and the script model 123.

[0034] A request for ETL processing generation is sent from the ETL worker terminal 20 to the information processing device 10 (step S1001). Subsequently, the information processing device 10 performs the following processing. That is, the mapping creation unit 13 creates mapping data by referring to the mapping model 121 and the vector DB 122 in accordance with the request (step S2001).

[0035] Here, the ETL worker terminal 20 is an input device for inputting ESG information before conversion, but the ESG information before conversion may be input from other terminals. Also, the ESG person terminal 30 is an output device for outputting ESG information after conversion, but the ESG information after conversion may be output from other terminals.

[0036] Next, the prompt creation unit 14 creates a prompt based on the mapping data (step S2002), and the ETL processing generation unit 15 generates ETL processing script data by referring to the script model 123 (step S2003).

[0037] Next, the ETL processing script verification unit 16 performs script verification on the ETL processing script data (step S2004). Then, the ESG format conversion unit 17 converts the ETL processing script data into ESG format based on the results, generating ESG information (step S2005), and this ESG information is provided to the ESG person's terminal 30 (step S2006).

[0038] On the ESG personnel terminal 30, ESG information provided by the information processing device 10 is displayed on a display not shown (step S3001).

[0039] Next, the specific processing procedure will be explained using Figures 5 to 7. Figures 5 to 7 are explanatory diagrams illustrating the processing procedure according to this embodiment. Following the sequence described above, the engineer first operates the engineer terminal 50 to input domain knowledge into the vector DB 122. After that, the ETL creator operates the ETL worker terminal 20 to request the information processing device 10 to generate the ETL process.

[0040] The information processing device 10 extracts items from both the company's proprietary format and the international standard format, and performs a vector database search on the related information according to the conversion rules of the mapping model 121.

[0041] In this case, as shown in Figure 5, the company's proprietary format 5001 defines electricity consumption and water consumption as 500 MWh and 10,000 cubic meters, respectively. The international standard format 5002 defines the conversion of the electricity consumption and water consumption from the company's proprietary format 5001 to total energy consumption and total water intake, respectively. Therefore, the units are converted from MWh to GJ, and cubic meters are treated as the same cubic meters without conversion.

[0042] Furthermore, in the vector database search, searching by electricity consumption retrieves information on total energy consumption, and searching by water usage retrieves information on total water intake.

[0043] Next, along with instructions to create prompts, the pre-conversion ESG information (items, data section), items in the international standard format, and RAG (supplementary information) are input into the generating AI, mapping model 121. While a detailed explanation of the prompts is omitted, the mapping data used to convert the ESG information is, for example, a CSV (Comma Separated Values) file.

[0044] In this way, mapping data is generated, and once prompts are created based on that mapping data, the ETL process is executed by referring to script model 123, as shown in Figure 6.

[0045] A detailed explanation of the script content in Figure 6 is omitted, but the processing steps are as follows: reading the source data (ESG information before conversion) CSV file, reading the mapping data CSV file, setting up a function to read the contents of the aforementioned CSV files, reading the source data and mapping data, setting up a unit conversion function, listing the converted data (ESG information) to be stored, executing the data conversion, generating the contents of the converted CSV file, displaying the results, and writing the converted data to the CSV file.

[0046] Then, as shown in Figure 7, the ESG information disclosure officer operates the ESG officer terminal 30, and based on the ETL processing script created as described above, the ESG information before conversion (company-specific format) is converted using the ESG conversion format, and the converted ESG information (international standard format) is obtained. In this way, the converted ESG information is displayed on a display (not shown) on the ESG officer terminal 30, and the ESG information disclosure officer can visually confirm the screen display of the ESG information.

[0047] Furthermore, using Figures 8 to 12, the process of converting ESG information from collection items to disclosure items will be explained in several patterns as an example. Figures 8 to 12 are explanatory diagrams illustrating specific examples of the processing according to this embodiment.

[0048] The above patterns are: Pattern 1: Collected items are aggregated and summarized to create a disclosure item; Pattern 2: Multiple collected items are processed using a specific program, summarized, and calculated to create a single disclosure item; Pattern 3A: Collected items do not become disclosure items and are therefore entered separately; Pattern 3B: Items that would become disclosure items if collected are already aggregated and therefore are entered separately; and Pattern 4: Collected items are used for purposes other than disclosure.

[0049] Regarding Pattern 1, a detailed explanation will be omitted, but as shown in Figure 8, the summary 1001 includes CO2 emissions as a collection item, a calculation method that calculates the total CO2 emissions by summing up the above collection items as a calculation process, and the calculation result as a disclosure item.

[0050] The conversion process 1002 follows this flow: ETL process generation request, mapping model generation, ETL process generation, ETL process verification, ETL process integration, and ESG format conversion.

[0051] Regarding Pattern 2, a detailed explanation will be omitted, but as shown in Figure 9, the summary 2001 includes energy consumption as a collection item, a calculation method that calculates total energy consumption by summing up the above collection items, and the calculation result as a disclosure item.

[0052] The conversion process 2002 follows this flow: ETL process generation request, mapping model generation, ETL process generation, ETL process verification, ETL process integration, and ESG format conversion.

[0053] Regarding Pattern 3A, a detailed explanation will be omitted here as well, but as shown in Figure 10, the summary 3001A includes the number of employees per department as a collection item, and "items that are not directly disclosed from the collection items but must be filled in separately" as disclosure items.

[0054] The conversion process 3002A follows this flow: ETL process generation request, mapping model generation, ETL process generation, ETL process verification, ETL process integration, and ESG format conversion.

[0055] Regarding Pattern 3B, a detailed explanation will be omitted here, but as shown in Figure 11, the summary 3001B includes the number of employees by region as a collection item and the number of employees by region as a disclosure item.

[0056] The conversion process 3002B follows this flow: ETL process generation request, mapping model generation, ETL process generation, ETL process verification, ETL process integration, and ESG format conversion.

[0057] Furthermore, although a detailed explanation of Pattern 4 will be omitted, as shown in Figure 12, the summary 4001 includes the following items for collection: energy consumption by department, and for purposes of use (for example, purposes of use other than disclosure purposes): "confirmation of energy consumption for internal audits," "evaluation of energy saving programs," and "energy efficiency improvement plans for each department."

[0058] The conversion process 4002 follows this flow: ETL process generation request, mapping model generation, ETL process generation, ETL process verification, ETL process integration, and ESG format conversion.

[0059] As described above, this embodiment makes it possible to facilitate the disclosure of ESG information. In particular, by supporting the creation of transformation rules automatically using machine learning, and by combining automatic rule generation with human modification, and generating ETL processing from the transformation rules, the transformation can be automated. As a result, users do not need to manually map the correspondences, and the effort required for ESG information disclosure is reduced.

[0060] Furthermore, it can adapt to the proliferation of standards and support conversion to formats compatible with each standard. In addition, SASB's version tracking function (format conversion support according to the version) allows processing to be done directly with the latest version, eliminating the hassle of differences between old and new versions.

[0061] Furthermore, the Corporate Sustainability Reporting Directive (CSRD) regulations will require companies to disclose ESG information at the same level as financial information. In other words, ESG information will need to be disclosed not at the company-wide level, but at the sector (business unit) level. Currently, for example, no one can answer the question, "What is the CO2 emission level of the financial business unit?" and if sectors are merged or consolidated, gathering information will become even more difficult, but this new regulation will be applicable to situations like this.

[0062] Furthermore, in the embodiment described above, the information processing system is formed by an information processing device 10, an ETL worker terminal 20, and an ESG person terminal 30, but it is also possible to configure it to incorporate a domain knowledge DB 60 into the storage unit 12.

[0063] Furthermore, each of the above-mentioned configurations, functional units, processing units, processing means, etc., may be implemented in hardware, in whole or in part, for example, by designing them as integrated circuits. Alternatively, each of the above-mentioned configurations, functions, etc., may be implemented in software by having the processor interpret and execute programs that realize each function. Information such as programs, tables, and files that realize each function can be stored in memory, hard disks, SSDs (Solid State Drives), or other recording devices, or in recording media such as IC cards, SD cards, or DVDs.

[0064] Furthermore, the arrangement of the various functional units, processing units, and databases described above is merely an example. The arrangement of the various functional units, processing units, and databases can be changed to the optimal arrangement from the standpoint of the performance, processing efficiency, and communication efficiency of the hardware and software of these devices.

[0065] Furthermore, the configuration of the database (schema, etc.) that stores the various types of data mentioned above can be flexibly modified from the perspective of efficient resource utilization, improved processing efficiency, improved access efficiency, and improved search efficiency. [Explanation of Symbols]

[0066] 10 Information Processing Devices 11 Communications Department 12 Storage section 13. Mapping Creation Department 14. Prompt Creation Section 15 ETL Processing Generation Unit 16 ETL Processing Script Verification Section 17 ESG format conversion section 20 ETL worker terminals 30 ESG personnel terminals 50 Technician terminals 60 Domain Knowledge Database 70 International Standard Databases 101 CPU 102 Programs 103 memory 104 Operating Devices 105 External storage device 106 Communication IF 107 Bus 121 Mapping Models 122 Vector DB 123 Script Model

Claims

1. An information processing device, A memory unit that stores a mapping model, vector data, and script model that have learned the correspondence between ESG (environmental, social, and governance) information and predetermined standards, A mapping creation unit that creates mapping data based on input ESG information by referring to the aforementioned mapping model and vector data, A prompt creation unit creates prompts based on the mapping data created by the mapping creation unit, A generation unit that generates an ETL (extract, transform, write) processing script based on the prompt created by the prompt creation unit, referring to the script model, A conversion unit that converts the ETL processing script generated by the generation unit into ESG format and generates ESG information, An information processing device characterized by comprising:

2. An information processing apparatus according to claim 1, further comprising a verification unit for verifying the ETL processing script generated by the generation unit, wherein the conversion unit converts the ETL processing script generated by the generation unit into ESG format based on the results verified by the verification unit.

3. The information processing apparatus according to claim 1, wherein the mapping creation unit generates mapping data according to a unit conversion rule between the original data item and the converted data item.

4. The information processing device according to claim 1, characterized in that the mapping model is a large-scale language model.

5. The information processing apparatus according to claim 1, characterized in that the script model is a large-scale language model.

6. An information processing device according to claim 1, wherein the device is connected to an external terminal in a manner that enables communication with the external terminal, and further has a communication unit that receives ESG information from the external terminal, and the mapping creation unit processes the ESG information received by the communication unit.

7. The information processing apparatus according to claim 1, wherein the conversion unit generates display ESG information.

8. An information processing device according to claim 7, characterized in that it is connected to an external device in a communicative manner and further has a communication unit that outputs display ESG information generated by the conversion unit to the external device.

9. An information processing device according to claim 1, characterized in that the predetermined standard is an international standard.

10. An information processing system, An input device for entering ESG (Environmental, Social, and Governance) information, An information processing device that converts ESG information input by the aforementioned input device according to predetermined standards, An output device that outputs ESG information converted by the aforementioned information processing device, Equipped with, The aforementioned information processing device is An information processing system characterized by using a pre-prepared learning model to convert ESG information input by the input device according to predetermined criteria to create a prompt, and generating ESG information according to the predetermined criteria using an ETL (extract, convert, write) processing script based on the prompt.

11. An information processing method that performs information processing using a computer, A mapping creation step involves referencing a mapping model and vector data stored in memory that have learned the correspondence between ESG (Environmental, Social, and Governance) information and predetermined standards, and creating mapping data based on the input ESG information. A prompt creation step that creates a prompt based on the mapping data created in the mapping creation step, A generation step that refers to the script model stored in the memory and generates an ETL (extract, transform, write) processing script based on the prompt created in the prompt creation step, A conversion step which converts the ETL processing script generated in the above generation step into ESG format to generate ESG information, An information processing method characterized by including

Images