Enterprise ESG data analysis method, device, equipment and medium

Abstract

The invention relates to an enterprise ESG data analysis method, device and equipment and a medium, and belongs to the technical field of data analysis, and the method comprises the steps: obtaining an ESG report file of a to-be-analyzed enterprise; acquiring an industry type corresponding to the to-be-analyzed enterprise; determining a knowledge graph corresponding to the to-be-analyzed enterprise based on the industry type; determining a to-be-extracted index based on the knowledge graph; extracting to-be-analyzed data corresponding to the to-be-extracted index from the ESG report file; generating a to-be-analyzed ERS evaluation file based on the to-be-extracted index and the to-be-analyzed data; and analyzing the to-be-analyzed enterprise based on the to-be-analyzed ERS evaluation file to obtain an analysis result. The method has the effect of improving the analysis efficiency and accuracy of the ESG report.

Description

Technical Field

[0001] This application relates to the technical field of data analysis, and in particular, to a method, device, equipment and medium for enterprise ESG data analysis. Background Art

[0002] ESG (Environmental, Social and Governance) evaluates the sustainability of an enterprise's operation and its impact on social values from three dimensions: environment, society and corporate governance. ESG itself is an evaluation concept based on value sustainability. Based on ESG evaluation, investors can observe the contributions of an enterprise in promoting economic sustainable development, fulfilling social responsibilities, etc.

[0003] When analyzing an enterprise's ESG report, generally, staff need to find the data of corresponding rating indicators in the ESG report file for ESG analysis. However, the number of evaluation indicators is often huge. It is inefficient and error-prone for staff to determine the data corresponding to the evaluation indicators in the enterprise's ESG report manually. Summary of the Invention

[0004] In order to improve the analysis efficiency and accuracy of ESG reports, this application provides a method, device, equipment and medium for enterprise ESG data analysis.

[0005] In a first aspect, this application provides a method for enterprise ESG data analysis, adopting the following technical solution:

[0006] A method for enterprise ESG data analysis includes:

[0007] Obtain the ESG report file of the enterprise to be analyzed;

[0008] Obtain the industry type corresponding to the enterprise to be analyzed;

[0009] Determine the knowledge graph corresponding to the enterprise to be analyzed based on the industry type;

[0010] Determine the indicators to be extracted based on the knowledge graph;

[0011] Extract the data to be analyzed corresponding to the indicators to be extracted from the ESG report file;

[0012] Generate an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed;

[0013] Analyze the enterprise to be analyzed based on the ERS evaluation file to be analyzed, and obtain the analysis result.

[0014] By adopting the above technical solution, a knowledge graph is determined based on the industry type, the indicators to be extracted are automatically determined through the knowledge graph, and the data to be analyzed corresponding to the indicators to be extracted is automatically extracted from the ESG report file. Compared with the way that staff manually determine the data to be analyzed corresponding to the indicators to be extracted in the enterprise ESG report, the analysis efficiency is improved to a certain extent. Moreover, using the knowledge graph to determine the indicators to be extracted and extract the data to be analyzed can ensure the accurate matching of data, avoid errors and omissions that may occur in the manual screening process, and improve the accuracy of data analysis.

[0015] Optionally, generating the to-be-analyzed ERS evaluation file based on the to-be-extracted indicators and the to-be-analyzed data includes:

[0016] Obtain the first code corresponding to the to-be-extracted indicator;

[0017] Obtain the second code corresponding to the first code;

[0018] Determine the adding position of the to-be-analyzed data in the file based on the second code, and add the to-be-analyzed data to the corresponding adding position;

[0019] When each to-be-extracted indicator corresponds to a to-be-analyzed data, obtain the to-be-analyzed ERS evaluation file;

[0020] Wherein, the first code is used to identify the position of the to-be-extracted indicator in the file, the second code is used to identify the position of the to-be-analyzed data in the file, and the first code and the second code are in one-to-one correspondence.

[0021] By adopting the above technical solution, the first code is used to identify the position of the to-be-extracted indicator in the file, and the second code is used to identify the position of the to-be-analyzed data in the file; through the first code and the second code, it is ensured that each to-be-extracted indicator can be accurately matched with the corresponding to-be-analyzed data, reducing the possibility of data misalignment or omission.

[0022] Optionally, extracting the to-be-analyzed data corresponding to the to-be-extracted indicator from the ESG report file includes:

[0023] Extract the to-be-analyzed data corresponding to the to-be-extracted indicator from the ESG report file based on the RAG model.

[0024] By adopting the above technical solution, the RAG model combines the advantages of information retrieval and generation models, can achieve efficient and accurate screening and processing of information, and in the ESG report file, the RAG model can quickly locate and extract data related to the to-be-extracted indicators, greatly reducing the time for manual reading, screening and extracting data, and improving the efficiency of data extraction.

[0025] Optionally, when the extraction of the data to be analyzed from the ESG report file is completed, the method further includes:

[0026] Query whether there are extraction metrics that do not include the data to be analyzed;

[0027] If so, extract the extraction metrics that do not include the data to be analyzed, and use the extracted extraction metrics as the information to be queried;

[0028] Search for query data that matches the information to be queried in the large database, and use the query data as the data to be analyzed.

[0029] By adopting the above technical solution, by querying and supplementing the missing data to be analyzed, the integrity of the data to be analyzed corresponding to all key metrics required for ESG assessment is ensured, thereby reducing the possibility of assessment deviation or omission caused by data missing.

[0030] Optionally, before using the query data as the data to be analyzed, it further includes:

[0031] Obtain the data quantity of the query data corresponding to each piece of information to be queried;

[0032] When the data quantity is greater than 1, obtain the data source and release time corresponding to each piece of query data;

[0033] Determine the first screening score and the second screening score corresponding to the data source and release time;

[0034] Calculate the total screening value based on the first screening score and the second screening score;

[0035] Based on the total screening value, determine a query data value from multiple pieces of query data as the data to be analyzed.

[0036] By adopting the above technical solution, when there are multiple pieces of query data, calculate the total screening value through the data source and release time information, comprehensively consider the authority of the data source and the timeliness of the release time, so as to select the optimal query data as the data to be analyzed, thereby improving the credibility of the report.

[0037] Optionally, the analysis of the enterprise to be analyzed based on the ERS evaluation file to be analyzed includes:

[0038] Obtain the weight value corresponding to each extraction metric and the analysis score corresponding to each data to be analyzed;

[0039] Calculate the ESG score of the current enterprise based on the analysis score and the corresponding weight value;

[0040] Compare the current enterprise's ESG score with the ESG scores of peer enterprises to obtain the performance level of the enterprise to be analyzed among its peers;

[0041] Compare the current enterprise's ESG score with the ESG score of the previous year to obtain the improvement level of the enterprise to be analyzed in terms of ESG;

[0042] Determine the ESG risks based on the indicators with lower or negative changes in the ESG score;

[0043] Generate an analysis report based on the performance level, improvement level, and ESG risks.

[0044] By adopting the above technical solution, by obtaining the weight value corresponding to each index to be extracted and the analysis score corresponding to each data to be analyzed, it is possible to comprehensively consider the performance of the enterprise in the three dimensions of environment, society, and corporate governance, ensuring the accuracy and comprehensiveness of the evaluation; comparing the current enterprise's ESG score with the ESG score of the previous year can reveal the improvement of the enterprise in terms of ESG and provide data support for the enterprise to formulate improvement measures. According to the indicators with lower or negative changes in the ESG score, the potential risks existing in the enterprise in terms of ESG can be accurately identified; through the identification and analysis of ESG risks, the enterprise can take measures in advance for risk prevention and control to avoid the negative impact of potential risks on the enterprise.

[0045] Optionally, before obtaining the weight value corresponding to each index to be extracted and the analysis score corresponding to each data to be analyzed, it further includes:

[0046] Query whether there is policy information corresponding to the index to be selected;

[0047] If so, determine the number of occurrences of the policy information and the source of publication corresponding to the policy information;

[0048] Determine the adjustment coefficient based on the number of occurrences and the source of publication;

[0049] Update the weight value of the index to be selected corresponding to the policy information based on the adjustment coefficient.

[0050] By adopting the above technical solution, by querying policy information and determining the adjustment coefficient based on the number of occurrences and the source of publication of the policy information, it is possible to dynamically adjust the weight values of each index in the ESG scoring model, making the weight distribution more reasonable and scientific. The introduction of policy information can guide enterprises to pay more attention to ESG compliance, ensure that enterprises actively respond to policy requirements during operation, and reduce potential policy risks.

[0051] In a second aspect, the present application provides an enterprise ESG data analysis device, adopting the following technical solution:

[0052] An enterprise ESG data analysis device, comprising:

[0053] A first acquisition module, configured to acquire an ESG report file of an enterprise to be analyzed;

[0054] A second acquisition module, configured to acquire the industry type corresponding to the enterprise to be analyzed;

[0055] A first determination module, configured to determine a knowledge graph corresponding to the enterprise to be analyzed based on the industry type;

[0056] A second determination module, configured to determine indicators to be extracted based on the knowledge graph;

[0057] An extraction module, configured to extract data to be analyzed corresponding to the indicators to be extracted from the ESG report file;

[0058] A generation module, configured to generate an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed;

[0059] An evaluation module, configured to analyze the enterprise to be analyzed based on the ERS evaluation file to be analyzed, and obtain an analysis result.

[0060] In a third aspect, the present application provides an electronic device, adopting the following technical solution:

[0061] An electronic device, comprising a processor and a memory, the processor being coupled to the memory;

[0062] The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of the first aspect.

[0063] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution:

[0064] A computer-readable storage medium, comprising a computer program or instruction, when the computer program or instruction runs on a computer, enabling the computer to execute the method according to any one of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a schematic flow chart showing an enterprise ESG data analysis method in an embodiment of the present application.

[0066] Figure 2 is a structural block diagram showing an enterprise ESG data analysis device in an embodiment of the present application.

[0067] Figure 3 is a structural block diagram showing an electronic device in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] The present application will be further described in detail below with reference to the accompanying drawings.

[0069] This specific embodiment is only an interpretation of the present application and does not limit the present application. After reading this specification, those skilled in the art can make modifications to this embodiment without creative contributions as needed, but as long as it is within the scope of the claims of the present application, it is protected by the patent law.

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without making creative efforts fall within the scope of protection of the present application.

[0071] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0072] The embodiments of the present application will be further described in detail below with reference to the drawings in the specification.

[0073] The embodiments of the present application provide a method for analyzing enterprise ESG data. This method for analyzing enterprise ESG data can be executed by an electronic device, which can be a server or a terminal device. The server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a desktop computer, etc., but is not limited thereto.

[0074] As Figure 1 shown, a method for analyzing enterprise ESG data, the main process of the method is described as follows (Steps S101 - S107):

[0075] Step S101, obtain the ESG report file of the enterprise to be analyzed;

[0076] In this embodiment, the user inputs the name of the enterprise that needs to analyze the ESG report into the electronic device through the input device of the electronic device, such as a keyboard, a mouse, and a touch control device. The electronic device uses the obtained enterprise name as the enterprise to be analyzed. The electronic device queries the ESG report file published by the enterprise to be analyzed. Among them, the ESG report files of each company can be imported into the electronic device in advance, or the electronic device can be used to query through the corresponding website. No specific limitation is made on this.

[0077] Step S102: Obtain the industry type corresponding to the enterprise to be analyzed;

[0078] Specifically, obtain the invoices and receipts sold by the enterprise to be analyzed; determine the industry type of the enterprise to be analyzed based on the invoices and receipts.

[0079] In this embodiment, when the electronic device obtains the enterprise name of the enterprise to be analyzed, it queries the financial data of the enterprise to be analyzed through the enterprise name. Among them, the financial data can be obtained through the internal enterprise system of the enterprise to be analyzed, or can be obtained through the corresponding website. No specific limitation is made on this. The financial data includes but is not limited to the invoices and receipts of the products sold by the enterprise to be analyzed. Determine the products sold by the enterprise to be analyzed through the invoices and receipts, determine the main business through the sold products, and thus determine the industry type of the enterprise to be analyzed.

[0080] Among them, the industry types include but are not limited to:

[0081] Energy and Utilities: Include utility companies such as electricity, gas, and water, and also include energy companies such as oil and natural gas; The main concerns of this type of industry in terms of ESG mainly include carbon emissions, energy consumption, energy efficiency, use of renewable energy, etc.

[0082] Materials and Industry: Include industrial companies such as mining, metals, chemicals, and building materials; The main concerns of this type of industry in terms of ESG mainly include resource consumption, environmental pollution, work safety, employee welfare, etc.

[0083] Consumer Staples and Discretionary Consumption: Include consumer goods companies such as food, beverages, tobacco, textiles, clothing, household appliances, and automobiles. The main concerns of this type of industry in terms of SG mainly include product quality, consumer rights protection, supply chain management, labor rights, etc.

[0084] Healthcare: Include healthcare companies such as pharmaceuticals, biotechnology, and medical devices; The main concerns of this type of industry in terms of ESG mainly include drug research and development, clinical trials, patient rights protection, medical waste treatment, etc.

[0085] Information Technology: covering information technology companies such as software, hardware, Internet, and communication; the main ESG concerns in this type of industry mainly include data security, privacy protection, network security, intellectual property, etc.

[0086] Financial Services: including financial companies such as banks, insurance, and securities. The main ESG concerns in this type of industry mainly include credit policies, investment strategies, risk management, customer privacy protection, etc.

[0087] Step S103: Determine the knowledge graph corresponding to the enterprise to be analyzed based on the industry type;

[0088] In this embodiment, since the attributes of different types of industries are different, the ESG concerns are also different. Therefore, different industry types correspond to different knowledge graphs. Among them, the knowledge graph is the ESG evaluation indicators that need to be analyzed for this type of industry. In this embodiment, the knowledge graph is collected by the staff according to industry standards to determine the evaluation indicators.

[0089] Table 1 shows the evaluation indicators required for a certain industry type.

[0090] Table 1

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] Step S104: Determine the indicators to be extracted based on the knowledge graph;

[0100] In this embodiment, the evaluation indicators in the knowledge graph are used as the indicators to be extracted, so as to extract the corresponding analysis data in the ESG report file.

[0101] Step S105: Extract the data to be analyzed corresponding to the indicators to be extracted from the ESG report file;

[0102] Specifically, the data to be analyzed corresponding to the indicators to be extracted in the ESG report file is extracted based on the RAG model.

[0103] In this embodiment, when obtaining an ESG report file, the file type of the ESG report file is first identified. The file types include, but are not limited to, PDF, word, and xlm. The OCR technology is used to convert the ESG report file into text data.

[0104] In this embodiment, the RAG model is used to determine the data to be analyzed corresponding to the metrics to be extracted in the ESG report file. The RAG model converts the text data in the ESG report file into a vector database, converts the metrics to be extracted into query vectors, and determines the data to be analyzed through the query vectors in the vector database.

[0105] Among them, the extraction content of the data to be analyzed includes, but is not limited to, the chapter and page number where the metric to be extracted is located, and the screenshot of the data to be analyzed corresponding to the metric to be extracted.

[0106] When the extraction of the data to be analyzed corresponding to all the metrics to be extracted is completed, the following content is also included:

[0107] Specifically, query whether there are metrics to be extracted that do not include the data to be analyzed; if so, extract the metrics to be extracted that do not include the data to be analyzed, and use the extracted metrics to be extracted as the information to be queried; search in the large database for the query data that matches the information to be queried, and use the query data as the data to be analyzed.

[0108] In this embodiment, the electronic device generates a query situation for the metrics to be extracted. The query situation includes whether the metrics to be extracted are disclosed in the ESG report file. When there is data to be analyzed corresponding to the extraction metrics in the ESG report file, it is determined that the data to be analyzed corresponding to the metrics to be extracted has been disclosed. When there is no data to be analyzed corresponding to the extraction metrics in the ESG report file, it is determined that the data to be analyzed corresponding to the metrics to be extracted has not been disclosed. After all the metrics to be extracted are extracted, query whether there are metrics to be extracted that are not disclosed, that is, metrics to be extracted that do not include the data to be analyzed. At this time, extract all the metrics to be extracted that do not include the data to be analyzed, use the extracted metrics to be extracted as the information to be queried, and use web crawler technology to query the query data that matches the information to be queried in the large database, so as to use the query data as the data to be analyzed.

[0109] Before using the query data as the data to be analyzed, the following is also included:

[0110] Obtain the data quantity of the query data corresponding to each piece of information to be queried; when the data quantity is greater than 1, obtain the data source and release time corresponding to each query data; determine the first screening score and the second screening score corresponding to the data source and release time; calculate the total screening score based on the first screening score and the second screening score; determine a query data value as the data to be analyzed from multiple query data based on the total screening score.

[0111] In this embodiment, when querying the information to be queried, there may be multiple query data for each query information. Among them, the multiple query data may come from different websites and different dates. Among them, the credibility of each data source is determined according to the retrieval times, evaluation quality, and whether it is an official channel of the query data on each website. Different first screening scores are determined according to the credibility, and the first screening scores determined by each data source are stored in the electronic device. There is a second screening score corresponding to the release time stored in the electronic device. Among them, the release time is divided into multiple time periods, and each time period corresponds to a second screening score. The data source and the release time are preset with different weight values. The screening total score is calculated according to the first screening score, the second screening score, and the corresponding weight values, and the query data with the largest screening total score is selected as the data to be analyzed.

[0112] In this embodiment, the data to be analyzed corresponding to the information to be queried is marked to distinguish it from the data to be analyzed in the ESG report file.

[0113] Step S106, generating a to-be-analyzed ERS evaluation file based on the to-be-extracted indicators and the data to be analyzed;

[0114] Specifically, obtaining the first code corresponding to the to-be-extracted indicator; obtaining the second code corresponding to the first code; determining the addition position of the data to be analyzed in the file based on the second code, and adding the data to be analyzed to the corresponding addition position; when each to-be-extracted indicator corresponds to a data to be analyzed, obtaining the to-be-analyzed ERS evaluation file; among them, the first code is used to identify the position of the to-be-extracted indicator in the file, the second code is used to identify the position of the data to be analyzed in the file, and the first code and the second code correspond one by one.

[0115] In this embodiment, when the to-be-extracted indicators are determined, a file template is obtained, and the to-be-extracted indicators are added to the corresponding positions of the file template. Each to-be-extracted indicator corresponds to a display area for the data to be analyzed, and the to-be-extracted indicators and the corresponding display areas for the data to be analyzed are bound in advance.

[0116] Specifically, obtaining the first position of the to-be-extracted indicator in the file, numbering the first position to obtain a first number; obtaining the second position of the to-be-analyzed data in the file, numbering the second position to obtain a second number; binding the to-be-extracted indicator and the to-be-analyzed data corresponding to the to-be-extracted indicator based on the first number and the second number;

[0117] In this embodiment, each position of the file template is numbered. Among them, the position of the to-be-extracted indicator is the first number, and the display area of the data to be analyzed is the second number, and the first number is bound to the corresponding second number.

[0118] When obtaining the data to be analyzed, obtain the first encoding of the extraction index corresponding to the data to be analyzed, determine the second encoding based on the first encoding, use the position corresponding to the second encoding as the addition position, and add the data to be analyzed to the corresponding addition position. When all extraction indexes correspond to a data to be analyzed, generate an ERS evaluation file to be analyzed.

[0119] Table 2 is the ERS evaluation file to be analyzed.

[0120] Table 2

[0121]

[0122] In this embodiment, when obtaining the ERS evaluation file to be analyzed, input the extraction index and the corresponding data to be analyzed into the verification model to verify whether the extraction index and the corresponding data to be analyzed match. When it is verified that the extraction index and the corresponding data to be analyzed do not match, generate an alarm message to correct the error data in a timely manner, thereby improving the accuracy of the ESG report analysis. Among them, the verification model can be a neural network model.

[0123] Step S107: Analyze the enterprise to be analyzed based on the ERS evaluation file to be analyzed to obtain an analysis result.

[0124] Specifically, obtain the weight value corresponding to each extraction index and the analysis score corresponding to each data to be analyzed; calculate the ESG score of the current enterprise based on the analysis score and the corresponding weight value; compare the ESG score of the current enterprise with the ESG scores of peer enterprises to obtain the performance level of the enterprise to be analyzed among its peers; compare the ESG score of the current enterprise with the ESG score of the previous year to obtain the improvement level of the enterprise to be analyzed in terms of ESG; determine the ESG risk according to the indicators with lower or negative changes in the ESG score; generate an analysis report based on the performance level, improvement level, and ESG risk.

[0125] In this embodiment, the weight value corresponding to each extraction index and the analysis score corresponding to the data to be analyzed are stored in the electronic device. When obtaining the ERS evaluation file to be analyzed, input the data to be analyzed into the recognition model to determine the analysis score corresponding to the data to be analyzed. The recognition model is a neural network model; according to the analysis scores of each extraction index and the corresponding weight values, use the weighted average method to calculate the ESG score of the current enterprise. The calculation formula is: ESG score = ∑(indicator analysis score × indicator weight value).

[0126] In this embodiment, the electronic device obtains similar enterprises of the same type, same scale, and same output as the enterprise to be analyzed. Among them, the similar enterprises can be determined according to the similarity algorithm, and no specific limitation is made here. In this embodiment, a chart form is used to intuitively display the performance level of the enterprise to be analyzed among the similar enterprises, such as a bar chart, a line chart, etc.

[0127] Compare the current enterprise's ESG score with the ESG score of the previous year to analyze the improvement level of the enterprise in terms of ESG, such as presenting it in a chart form.

[0128] According to the indicators with lower or negative changes in the ESG score, determine the ESG risks faced by the enterprise. Among them, the risks include but are not limited to environmental pollution, lack of social responsibility, and imperfect governance structure; conduct in-depth analysis of the identified ESG risks, evaluate their impact on aspects such as the enterprise's finance, reputation, and operation, and can also determine the priority of the risks in order to formulate targeted risk management strategies.

[0129] Before obtaining the weight value corresponding to each index to be extracted and the analysis score corresponding to each data to be analyzed, the following also includes:

[0130] Specifically, query whether there is policy information corresponding to the index to be selected; if so, determine the number of occurrences of the policy information and the corresponding release source of the policy information; determine the adjustment coefficient based on the number of occurrences and the release source; update the weight value corresponding to the index to be selected corresponding to the policy information based on the adjustment coefficient.

[0131] In this embodiment, count the number of occurrences of the policy information related to each index to be selected. The more the number of occurrences, the higher the importance of the index at the policy level. And different release sources may have different influences and authorities, so it is necessary to evaluate according to the level and influence of the release source.

[0132] In this embodiment, determine the adjustment coefficient of each index to be selected based on the number of occurrences of the policy information and the release source. The adjustment coefficient is a value between 0 and 1, which is used to reflect the influence degree of the policy information and the number of occurrences on the index weight value. Query the reference weight value corresponding to the number of occurrences and the release source in the electronic device, obtain the reference score corresponding to the release source, calculate the adjustment score according to the number of occurrences, the weight corresponding to the number of occurrences, the reference score, and the reference weight value corresponding to the release source, determine the adjustment coefficient according to the adjustment score. Among them, there is a preset adjustment score interval in the electronic device, and each adjustment score interval corresponds to an adjustment coefficient. When the adjustment score is calculated, determine the adjustment score interval where the adjustment score is located, so as to determine the corresponding adjustment coefficient. The calculation formula for the updated weight value can be: updated weight value = original weight value × adjustment coefficient.

[0133] In this embodiment, after updating the weight value, the original weight value is replaced to ensure that the enterprise actively responds to policy requirements during operation and reduces potential policy risks.

[0134] It should be noted that the occurrence times, the reference weight value corresponding to the release source, and the reference score corresponding to the release source are stored in the electronic device in advance. The reference score is used to evaluate the weight of different release sources. For example, government documents may obtain the highest score, while social media may obtain a lower score; the occurrence times refer to the reference weight value, and different weight values are set according to the number of occurrence times. For example, the more the occurrence times, the higher the weight value (but it should be noted to avoid linear growth to prevent over-amplifying the influence of certain indicators); the release source refers to the reference weight value, and different weight values are set according to the credibility and influence of the release source. For example, policy information released by government documents and authoritative institutions may have a higher weight value, while policy information released by social media, personal blogs, etc. may have a lower weight value.

[0135] Figure 2 It is a structural block diagram of an enterprise ESG data analysis device 200 provided by this application. As Figure 2 shown, the enterprise ESG data analysis device 200 mainly includes:

[0136] The first acquisition module 201 is used to acquire the ESG report file of the enterprise to be analyzed;

[0137] The second acquisition module 202 is used to acquire the industry type corresponding to the enterprise to be analyzed;

[0138] The first determination module 203 is used to determine the knowledge graph corresponding to the enterprise to be analyzed based on the industry type;

[0139] The second determination module 204 is used to determine the indicators to be extracted based on the knowledge graph;

[0140] The extraction module 205 is used to extract the data to be analyzed corresponding to the indicators to be extracted from the ESG report file;

[0141] The generation module 206 is used to generate an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed;

[0142] The evaluation module 207 is used to analyze the enterprise to be analyzed based on the ERS evaluation file to be analyzed and obtain the analysis result.

[0143] As an optional implementation manner in this embodiment, the generation module 206 is specifically used for:

[0144] Obtain the first code corresponding to the index to be extracted; obtain the second code corresponding to the first code; determine the addition position of the data to be analyzed in the file based on the second code, and add the data to be analyzed to the corresponding addition position; when each index to be extracted corresponds to a data to be analyzed, obtain the ERS evaluation file to be analyzed; wherein, the first code is used to identify the position of the index to be extracted in the file, the second code is used to identify the position of the data to be analyzed in the file, and the first code and the second code are in one-to-one correspondence.

[0145] As an optional implementation manner in this embodiment, the extraction module 205 is specifically configured to:

[0146] Extract the data to be analyzed corresponding to the index to be extracted in the ESG report file based on the RAG model.

[0147] As an optional implementation manner in this embodiment, the enterprise ESG data analysis device 200 further includes:

[0148] A query module, configured to query whether there is an index to be extracted that does not include the data to be analyzed when the extraction of the data to be analyzed from the ESG report file ends; if so, extract the index to be extracted that does not include the data to be analyzed, and use the extracted index to be extracted as the information to be queried;

[0149] A search module, configured to search for query data matching the information to be queried in the large database, and use the query data as the data to be analyzed.

[0150] As an optional implementation manner in this embodiment, the enterprise ESG data analysis device 200 further includes:

[0151] A quantity acquisition module, configured to acquire the data quantity of the query data corresponding to each piece of information to be queried before using the query data as the data to be analyzed;

[0152] A time acquisition module, configured to acquire the data source and release time of each query data when the data quantity is greater than 1;

[0153] A score determination module, configured to determine a first screening score and a second screening score corresponding to the data source and release time;

[0154] A calculation module, configured to calculate a screening total value based on the first screening score and the second screening score;

[0155] A data determination module, configured to determine a query data value as the data to be analyzed from multiple query data based on the screening total value.

[0156] As an optional implementation manner in this embodiment, the evaluation module 207 is specifically configured to:

[0157] Obtain the weight value corresponding to each index to be extracted and the analysis score corresponding to each data to be analyzed; calculate the ESG score of the current enterprise based on the analysis score and the corresponding weight value; compare the ESG score of the current enterprise with the ESG scores of peer enterprises to obtain the performance level of the enterprise to be analyzed among its peers; compare the ESG score of the current enterprise with the ESG score of the previous year to obtain the improvement level of the enterprise to be analyzed in terms of ESG; determine the ESG risk according to the index with a lower or negative change in the ESG score; generate an analysis report based on the performance level, improvement level, and ESG risk.

[0158] As an alternative implementation manner of this embodiment, the enterprise ESG data analysis device 200 further includes:

[0159] A policy information query module, configured to query whether there is policy information corresponding to the index to be selected before obtaining the weight value corresponding to each index to be extracted and the analysis score corresponding to each data to be analyzed; if so, determine the number of occurrences of the policy information and the release source corresponding to the policy information;

[0160] A coefficient determination module, configured to determine an adjustment coefficient based on the number of occurrences and the release source;

[0161] An update module, configured to update the weight value corresponding to the index to be selected corresponding to the policy information based on the adjustment coefficient;

[0162] A score calculation module, configured to calculate the ESG score of the current enterprise based on the updated weight value.

[0163] Each functional module in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part. If the functions are implemented in the form of software functional modules and sold or used as an independent product, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of a method for analyzing enterprise ESG data in various embodiments of the present application.

[0164] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and module can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0165] Figure 3 This is a structural block diagram of an electronic device 300 provided in an embodiment of the present application. AsFigure 3 As shown, the electronic device 300 includes a memory 301, a processor 302, and a communication bus 303; the memory 301 and the processor 302 are connected via the communication bus 303. Stored on the memory 301 is an enterprise ESG data analysis method that can be loaded and executed by the processor 302 as provided in the above embodiments.

[0166] The memory 301 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 301 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function, and instructions for implementing an enterprise ESG data analysis method provided in the above embodiments, etc.; the data storage area can store data involved in an enterprise ESG data analysis method provided in the above embodiments, etc.

[0167] The processor 302 may include one or more processing cores. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 301, the processor 302 calls data stored in the memory 301 and executes various functions of this application and processes data. The processor 302 can be at least one of an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, and a microprocessor. It can be understood that for different devices, the electronic devices for implementing the functions of the above processor 302 can also be others, and the embodiments of this application do not make specific limitations.

[0168] The communication bus 303 may include a path for transmitting information between the above components. The communication bus 303 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 303 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 3 only a double arrow is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0169] An embodiment of the present application provides a computer-readable storage medium storing a computer program that can be loaded and executed by a processor to implement an enterprise ESG data analysis method as provided in the foregoing embodiment.

[0170] In this embodiment, the computer-readable storage medium may be a tangible device that holds and stores instructions used by an instruction execution device. The computer-readable storage medium may be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the foregoing. Specifically, the computer-readable storage medium may be a portable computer disk, a hard disk, a USB flash drive, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a podium random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, and any combination of the foregoing.

[0171] The term "including", "comprising", or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or device.

[0172] The above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solution formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the foregoing application concept. For example, a technical solution formed by mutually replacing the above features with (but not limited to) technical features having similar functions in the present application.

Claims

1. A method for analyzing enterprise ESG data, characterized in that: include: Obtain ESG report documents of the companies to be analyzed; Obtaining the industry type corresponding to the enterprise to be analyzed; Determine the knowledge graph corresponding to the enterprise to be analyzed based on the industry type; Determine the index to be extracted based on the knowledge graph; Extracting the data to be analyzed corresponding to the indicator to be extracted from the ESG report file; Generate an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed; The enterprise to be analyzed is analyzed based on the ERS evaluation file to be analyzed to obtain an analysis result.

2. The method according to claim 1, characterized in that The step of generating an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed includes: Obtaining a first code corresponding to the indicator to be extracted; Obtain a second code corresponding to the first code; Determine the adding position of the data to be analyzed in the file based on the second code, and add the data to be analyzed to the corresponding adding position; When each of the indicators to be extracted corresponds to a piece of data to be analyzed, the ERS evaluation file to be analyzed is obtained; The first code is used to identify the position of the indicator to be extracted in the file, and the second code is used to identify the position of the data to be analyzed in the file, and the first code and the second code correspond one to one.

3. The method according to claim 1, characterized in that Extracting the data to be analyzed corresponding to the indicator to be extracted from the ESG report file includes: The data to be analyzed corresponding to the indicators to be extracted in the ESG report file are extracted based on the RAG model.

4. The method according to claim 3, characterized in that When the extraction of the data to be analyzed from the ESG report file is completed, the method further includes: Check whether there are any indicators to be extracted that do not include the data to be analyzed; If so, extract the indicators to be extracted that do not include the data to be analyzed, and use the extracted indicators to be extracted as the information to be queried; Search a large database for query data that matches the information to be queried, and use the query data as the data to be analyzed.

5. The method according to claim 4, characterized in that Before using the query data as the data to be analyzed, the method further includes: Obtain the data quantity of the query data corresponding to each of the information to be queried; When the number of data is greater than 1, obtain the data source and release time corresponding to each query data; Determine a first screening score and a second screening score corresponding to the data source and release time; Calculate a total screening value based on the first screening score and the second screening score; A query data value is determined from the plurality of query data based on the total screening value as the data to be analyzed.

6. The method according to claim 1, characterized in that The analyzing the enterprise to be analyzed based on the ERS evaluation file to be analyzed includes: Obtaining a weight value corresponding to each of the indicators to be extracted and an analysis score corresponding to each of the data to be analyzed; Calculate the current enterprise ESG score based on the analysis score and the corresponding weight value; Compare the ESG score of the current enterprise with the ESG scores of peer enterprises to obtain the performance level of the enterprise to be analyzed among its peers; Compare the current enterprise ESG score with the ESG score of the previous year to obtain the improvement level of the enterprise to be analyzed in terms of ESG; Identify ESG risks based on indicators of lower or negative changes in ESG scores; Generate an analysis report based on the performance level, improvement level and ESG risks.

7. The method according to claim 6, characterized in that Before obtaining the weight value corresponding to each of the to-be-extracted indicators and the analysis score corresponding to each of the to-be-analyzed data, the method further includes: Query whether there is policy information corresponding to the indicator to be selected; If yes, determining the number of times the policy information appears and the corresponding publishing source of the policy information; determining an adjustment factor based on the number of times and the source of the publication; The weight value corresponding to the to-be-selected indicator corresponding to the policy information is updated based on the adjustment coefficient.

8. An enterprise ESG data analysis device, characterized in that: include: The first acquisition module is used to obtain the ESG report file of the enterprise to be analyzed; The second acquisition module is used to acquire the industry type corresponding to the enterprise to be analyzed; A first determination module is used to determine the knowledge graph corresponding to the enterprise to be analyzed based on the industry type; A second determination module, used to determine the index to be extracted based on the knowledge graph; An extraction module, used to extract the data to be analyzed corresponding to the to-be-extracted indicator from the ESG report file; A generating module, used for generating an ERS evaluation file to be analyzed based on the indicators to be extracted and the data to be analyzed; The evaluation module is used to analyze the enterprise to be analyzed based on the ERS evaluation file to be analyzed to obtain an analysis result.

9. An electronic device, characterized in that: comprising a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The method comprises a computer program or an instruction, which, when executed on a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images