ESG rating method and device, electronic equipment and computer readable storage medium

Abstract

The embodiment of the invention provides an ESG rating method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of ESG rating, and the method comprises the steps: collecting multiple pieces of enterprise performance data from different data sources, the enterprise performance data being performance data of a to-be-evaluated enterprise in the aspects of environment, society and governance, and the enterprise performance data being performance data of the to-be-evaluated enterprise in the aspects of environment, society and governance; and performing standardization processing on the plurality of enterprise performance data to obtain preliminary enterprise performance data, performing feature extraction on the preliminary enterprise performance data to obtain performance features, and inputting the performance features into the trained ESG rating model to obtain an ESG rating result of the to-be-evaluated enterprise. According to the method, the data of various data sources are automatically collected, the integrity of the data is improved, standardization processing is carried out on the obtained enterprise performance data, the consistency and accuracy of the data are improved, and therefore the accuracy of an ESG rating result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ESG rating, and in particular to an ESG rating method, device, electronic device and computer-readable storage medium. Background Art

[0002] With the world's increasing emphasis on sustainable development, ESG ratings have become an important tool for measuring a company's social responsibility and sustainable development capabilities. ESG ratings refer to the evaluation of a company from three dimensions: environment, social responsibility, and governance.

[0003] Traditional ESG rating methods have problems in data acquisition and processing, such as incomplete data collection and uneven data quality, which leads to low accuracy of ESG rating results. Summary of the invention

[0004] The object of the present invention is to provide an ESG rating method, device, electronic device and computer-readable storage medium to solve the problem of inaccurate ESG rating results.

[0005] In a first aspect, the present invention provides an ESG rating method, the method comprising:

[0006] Collecting multiple enterprise performance data from different data sources; the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance;

[0007] Standardizing the plurality of enterprise performance data to obtain preliminary enterprise performance data;

[0008] Performing feature extraction on the preliminary enterprise performance data to obtain performance features;

[0009] The performance characteristics are input into a trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated; wherein the trained ESG rating model is trained based on the historical performance characteristics of multiple enterprises and the ESG rating results of each of the enterprises.

[0010] In an optional implementation manner, the step of performing standardization processing on the plurality of enterprise performance data to obtain preliminary enterprise performance data includes:

[0011] Preprocessing the plurality of enterprise performance data to obtain a plurality of preprocessed enterprise performance data;

[0012] Deeply cleaning the plurality of pre-processed enterprise performance data;

[0013] The multiple pre-processed enterprise performance data after deep cleaning are merged to obtain the preliminary enterprise performance data.

[0014] In an optional implementation manner, the preprocessing the plurality of enterprise performance data to obtain a plurality of preprocessed enterprise performance data includes:

[0015] Conduct preliminary cleaning of the performance data of multiple companies;

[0016] The plurality of enterprise performance data after preliminary cleaning are converted into a unified format to obtain the plurality of pre-processed enterprise performance data.

[0017] In an optional embodiment, the performance feature includes a plurality of performance evaluation features, and the step of inputting the performance feature into a trained ESG rating model to obtain the ESG rating result of the enterprise to be evaluated includes:

[0018] Calculating a score corresponding to each of the performance assessment characteristics;

[0019] According to the preset weight and score corresponding to each performance evaluation feature, the ESG rating result of the enterprise to be evaluated is obtained.

[0020] In an optional embodiment, the method further comprises:

[0021] A result explanation interface is generated according to the relationship between each of the performance evaluation characteristics and the ESG rating result of the enterprise to be evaluated; the result explanation interface is used to display the impact of each of the performance evaluation characteristics on the ESG rating result of the enterprise to be evaluated.

[0022] In an optional embodiment, the trained ESG rating model is trained by the following steps:

[0023] Obtain training samples and annotation results corresponding to the training samples: wherein the training samples are the historical performance characteristics of each of the enterprises, and the annotation results are the ESG rating results of each of the enterprises;

[0024] Inputting the training sample into a pre-established machine learning model to obtain a predicted ESG rating result corresponding to the training sample;

[0025] The parameters of the machine learning model are iteratively updated according to the predicted ESG rating results corresponding to the training samples and the annotation results to obtain a trained ESG rating model.

[0026] In an optional implementation, after obtaining the ESG rating result of the enterprise to be evaluated, the method further includes:

[0027] The ESG rating result of the enterprise to be evaluated is sent to the management system of the enterprise to be evaluated to provide management auxiliary information for the enterprise to be evaluated.

[0028] In a second aspect, the present invention provides an ESG rating device, the device comprising:

[0029] An enterprise performance data acquisition module is used to collect a plurality of enterprise performance data from different data sources; the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance;

[0030] An enterprise performance data processing module, used for performing standardization processing on the plurality of enterprise performance data to obtain preliminary enterprise performance data;

[0031] A performance feature extraction module, used to extract features from the preliminary enterprise performance data to obtain performance features;

[0032] The ESG rating result acquisition module is used to input the performance characteristics into the trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated; wherein the trained ESG rating model is trained based on the historical performance characteristics of multiple enterprises and the ESG rating results of each of the enterprises.

[0033] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and the computer program executable by the processor is used to implement the ESG rating method described in any of the aforementioned embodiments.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the ESG rating method as described in any one of the aforementioned embodiments is implemented.

[0035] The ESG rating method, device, electronic device and computer-readable storage medium provided by the embodiment of the present invention include: collecting multiple enterprise performance data from different data sources, the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance, standardizing the multiple enterprise performance data to obtain preliminary enterprise performance data, extracting features from the preliminary enterprise performance data to obtain performance features, inputting the performance features into the trained ESG rating model, and obtaining the ESG rating results of the enterprise to be evaluated, wherein the trained ESG rating model is trained based on the historical performance features of multiple enterprises and the ESG rating results of each enterprise. By automatically collecting data from multiple data sources, the integrity of the data is improved, and the obtained enterprise performance data is standardized to improve the consistency and accuracy of the data, thereby improving the accuracy of the ESG rating results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A schematic diagram of a process of an ESG rating method provided by an embodiment of the present invention is shown;

[0038] Figure 2 A schematic diagram of a process of obtaining a trained ESG rating model provided by an embodiment of the present invention is shown;

[0039] Figure 3 Another schematic diagram of a flow chart of an ESG rating method provided by an embodiment of the present invention is shown;

[0040] Figure 4 Another schematic diagram of a flow chart of an ESG rating method provided by an embodiment of the present invention is shown;

[0041] Figure 5 A module block diagram of an ESG rating device provided by an embodiment of the present invention is shown;

[0042] Figure 6 A block diagram of an electronic device provided by an embodiment of the present invention is shown.

[0043] Icons: 100 - electronic device; 110 - memory; 120 - processor; 130 - communication module; 200 - ESG rating device; 210 - enterprise performance data acquisition module; 220 - enterprise performance data processing module; 230 - performance feature extraction module; 240 - ESG rating result acquisition module. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0045] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0047] See also Figure 1 , Figure 1 A schematic diagram of a process flow of an ESG rating method provided by an embodiment of the present invention is shown. The ESG rating method can be applied to electronic devices. The specific process of this embodiment is described below using electronic devices as an example. Figure 1 The process shown in the figure is described in detail. The ESG rating method can specifically include the following steps:

[0048] Step 110: Collect multiple enterprise performance data from different data sources.

[0049] Among them, enterprise performance data refers to the performance data of the enterprise to be evaluated in terms of environment, society and governance. Data sources may include information disclosed by the enterprise, third-party databases and information released by industry organizations. Specifically, information disclosed by the enterprise may include annual reports, social responsibility reports, environmental reports, supply chain data and customer feedback; data provided by third parties may include media reports; information released by industry organizations may include environmental protection rules, industry labor standards, etc.

[0050] In some embodiments, the electronic device stores data types of corporate performance data to be collected, and the electronic device can determine the corresponding data source according to the data type, and collect multiple corporate performance data from different data sources. The data types may include environmental indicator types, social contribution types, and corporate governance structure types.

[0051] For example, the electronic device can search for environmental reports from the company's public information based on the type of environmental indicators, determine the environmental report as a data source, collect environmental performance data of the company to be evaluated from the environmental report, and use the collected performance data as the company's performance data.

[0052] In some embodiments, the electronic device may use a machine learning algorithm to predict and recommend data sources that have a higher impact on ESG rating results, and then collect corporate performance data from the data sources.

[0053] In some implementations, the electronic device may collect enterprise performance data from multiple determined data sources through an automatic crawling tool, and the electronic device may also collect enterprise performance data from a third-party database through an API (Application Programming Interface). The automatic crawling tool may include a crawler tool and a simple data collector.

[0054] In some implementations, the electronic device may collect multiple enterprise performance data from different data sources in real time to ensure the timeliness of the enterprise performance data, thereby ensuring the timeliness of the rating results.

[0055] Step 120: Standardize the multiple enterprise performance data to obtain preliminary enterprise performance data.

[0056] Step 130: Extract features from the preliminary enterprise performance data to obtain performance features.

[0057] In some embodiments, the electronic device can perform text classification, sentiment analysis, entity recognition, etc. on the preliminary enterprise performance data through natural language processing (NLP) technology, extract data reflecting enterprise performance from the preliminary enterprise performance data, and perform feature extraction on the data processed by the NLP technology to obtain performance features. The performance features may be gas emissions, employee satisfaction, etc.

[0058] Step 140: Input the performance features into the trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated.

[0059] Among them, the trained ESG rating model is trained based on the historical performance characteristics of multiple companies and the ESG rating results of each company.

[0060] In some embodiments, the ESG rating result of the enterprise to be evaluated may be a numerical value, for example, the ESG rating result of the enterprise to be evaluated is 89; the ESG rating result of the enterprise to be evaluated is a grade, for example, the ESG rating result of the enterprise to be evaluated is A.

[0061] In some embodiments, the electronic device can extract features from the historical data of the corresponding enterprise, obtain the historical performance features of the corresponding enterprise, and use the historical performance features of each enterprise as training samples. Experts can evaluate each enterprise based on the historical performance features of each enterprise and predetermined rating standards, and label each enterprise based on the evaluation results to obtain the ESG rating results of each enterprise.

[0062] The ESG rating method provided by the embodiment of the present invention collects multiple enterprise performance data from different data sources, the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance, standardizes the multiple enterprise performance data, obtains preliminary enterprise performance data, extracts features from the preliminary enterprise performance data, obtains performance features, inputs the performance features into the trained ESG rating model, and obtains the ESG rating results of the enterprise to be evaluated, wherein the trained ESG rating model is trained based on the historical performance features of multiple enterprises and the ESG rating results of each enterprise. By automatically collecting data from multiple data sources, the integrity of the data is improved, and the obtained enterprise performance data is standardized to improve the consistency and accuracy of the data, thereby improving the accuracy of the ESG rating results.

[0063] In some embodiments, the historical performance characteristics of each enterprise are used as input parameters, and the ESG rating results of each enterprise are used as reference output parameters to train the pre-established machine learning model. Figure 2 , Figure 2 A flowchart of a process for obtaining a trained ESG rating model provided by an embodiment of the present invention is shown.

[0064] Step 210: Obtain training samples and annotation results corresponding to the training samples.

[0065] The training samples are the historical performance characteristics of each company, and the labeled results are the ESG rating results of each company.

[0066] In some embodiments, the historical performance characteristics of a company may be performance characteristics extracted from company performance data previously used in ESG ratings.

[0067] Step 220: Input the training samples into a pre-established machine learning model to obtain the predicted ESG rating results corresponding to the training samples.

[0068] Among them, the pre-established machine learning models may include regression models, classification models, neural networks, etc.

[0069] In some embodiments, a corresponding machine learning model can be selected according to the nature of the task of obtaining ESG rating results.

[0070] As an implementation method, the nature of the task of obtaining the ESG rating results is a regression task, that is, the ESG rating results are to predict a continuous numerical value, for example, the ESG rating score, and a regression model can be used, such as linear regression, decision tree regression, support vector regression (SVR) and neural network, etc. The neural network can be LSTM (Long Short-Term Memory) and GRU (Gated Recurrent) etc.

[0071] As another implementation, the nature of the task of obtaining ESG rating results is a classification task, that is, the ESG rating results are levels. For example, the ESG rating results are divided into several levels (such as A, B, and C). Then, classification models are used, such as logistic regression, decision trees, random forests, gradient boosting machines (GBM), support vector machines (SVM), and classification neural networks.

[0072] Step 230: Iteratively update the parameters of the machine learning model according to the predicted ESG rating results and annotation results corresponding to the training samples to obtain a trained ESG rating model.

[0073] In some embodiments, after the training sample is input into a pre-established machine learning model, the pre-established machine learning model will output the corresponding predicted ESG rating results, compare the predicted ESG rating results with the actual labeled results (that is, the ESG rating results of the companies corresponding to the training samples), calculate the loss function and then use the optimization algorithm to minimize the loss function, and iteratively update the parameters of the pre-established machine learning model until the pre-established machine learning model reaches the stopping condition that is satisfied, thereby obtaining a trained ESG rating model.

[0074] The electronic device may calculate the loss function by means of square error, cross entropy loss, etc. The optimization algorithm may include gradient descent and Adam optimizer, etc. The stop condition may include the maximum number of iterations and the loss function change amount being less than a preset change amount, etc.

[0075] In some embodiments, the method of iteratively updating the parameters of the pre-established machine learning model may include grid search, random search, and Bayesian optimization.

[0076] In some embodiments, the electronic device may adjust the trained ESG rating model based on user feedback on the ESG rating results of the enterprise to be evaluated to adapt to changes in the rating standards.

[0077] To improve data quality, Figure 3As shown, step 120 specifically further includes the following steps:

[0078] Step 121: Preprocess a plurality of enterprise performance data to obtain a plurality of preprocessed enterprise performance data.

[0079] In some implementations, the electronic device may perform preliminary cleaning on the plurality of enterprise performance data, convert the plurality of enterprise performance data after preliminary cleaning into a unified format, and obtain a plurality of pre-processed enterprise performance data.

[0080] In some implementations, the electronic device removes duplicate items from the plurality of enterprise performance data, checks for abnormal values ​​in the plurality of enterprise performance data, and corrects or deletes the abnormal values ​​to complete preliminary cleaning of the plurality of enterprise performance data. The abnormal values ​​may include negative income, non-numeric dates, etc.

[0081] In some implementations, the electronic device may convert text data in the enterprise performance data into a unified expression and a unified format, and convert digital data in the enterprise performance data into a unified format and a unified unit.

[0082] Step 122: Deep cleansing of the plurality of pre-processed enterprise performance data.

[0083] In some embodiments, the plurality of pre-processed enterprise performance data are checked for consistency, such as whether the date format is consistent, and whether the profit and income in the financial data are consistent. The format of each pre-processed enterprise performance data is also checked to see whether it is correct, such as date, text, and number.

[0084] In some embodiments, multiple pre-processed enterprise performance data can be filled in. If the pre-processed enterprise performance data is missing too much data or is incomplete, the pre-processed enterprise performance data will be deleted; prediction models such as mean model, median model, mode model, etc. can be used to fill in the pre-processed enterprise performance data; if the data type of the pre-processed enterprise performance data meets the preset non-missing type, the preset filling value is used to fill the missing value.

[0085] Step 123: A plurality of pre-processed enterprise performance data after deep cleaning are integrated to obtain preliminary enterprise performance data.

[0086] In some implementations, the multiple pre-processed enterprise performance data after deep cleaning are merged by a data merging method to obtain preliminary enterprise performance data. The data merging method may include a JOIN operation of SQL, a merge method of pandas, and the like.

[0087] It can be understood that by preprocessing multiple corporate performance data, obtaining multiple preprocessed corporate performance data, and deeply cleaning multiple preprocessed corporate performance data, high-quality and complete preliminary corporate performance data can be obtained, providing a solid foundation for subsequent analysis, thereby improving the accuracy of ESG rating results.

[0088] Below, the process of obtaining ESG rating results based on the trained ESG rating model is described in detail. Performance features include multiple performance evaluation features, see Figure 4 , the above step 140 may specifically include the following steps:

[0089] Step 141: Calculate the score corresponding to each performance evaluation feature.

[0090] For example, the full score for each performance evaluation feature is 10 points, and the score is determined based on the performance of the enterprise to be evaluated relative to the industry average. The performance evaluation features include that the energy utilization efficiency of the enterprise to be evaluated is 20% higher than the industry average, and the waste recycling rate of the enterprise to be evaluated is 15% higher than the industry average. The energy utilization efficiency of the enterprise to be evaluated is 20% higher than the industry average, and the corresponding score is 9 points; the waste recycling rate of the enterprise to be evaluated is 15% higher than the industry average, and the corresponding score is 8 points.

[0091] Step 142: Obtain the ESG rating result of the enterprise to be evaluated based on the preset weight and score corresponding to each performance evaluation feature.

[0092] For example, the performance evaluation characteristics include working conditions and employee benefits as well as data privacy and network security. The preset weights for working conditions and employee benefits are 60%, and the preset weights for data privacy and network are 40%. The scores for working conditions and employee benefits are 8, and the scores for data privacy and network are 6. The ESG rating result of the company to be evaluated is 7.2 (8×60%+6×40%=7.2).

[0093] In order to enhance the transparency and comprehensibility of ESG rating results, the above-mentioned ESG rating method also includes: generating a result explanation interface based on the relationship between each performance evaluation feature and the ESG rating result of the enterprise to be evaluated; the result explanation interface is used to display the impact of each performance evaluation feature on the ESG rating result of the enterprise to be evaluated.

[0094] In some embodiments, the electronic device can generate a layer-by-layer parsing diagram during the process of the trained ESG rating model generating the ESG rating results of the enterprise to be evaluated, and the layer-by-layer parsing diagram is used to show the processing process of the performance evaluation features in the ESG rating model from the input layer to the output layer. The result interpretation interface can include a layer-by-layer parsing diagram.

[0095] If the trained ESG rating model is a decision tree model, the corresponding decision tree is directly visualized to obtain a layer-by-layer analysis diagram to display the decision logic of the trained ESG rating model.

[0096] If the trained ESG rating model is a neural network, rule extraction technology is used to interpret the behavior of the trained ESG rating model into a set of understandable rules, which are then displayed through a layer-by-layer analysis diagram.

[0097] In some embodiments, the electronic device can generate a simulation example based on the relationship between each performance evaluation feature and the ESG rating result of the enterprise to be evaluated, showing how the predicted result will change when the specific feature value changes. The electronic device can also generate a counter-simulation example based on the relationship between each performance evaluation feature and the ESG rating result of the enterprise to be evaluated. The counterfactual simulation example is used to show the conditions required to achieve different ESG rating results. For example, if the ESG rating result is to be improved to A level, the counter-simulation example can show which performance evaluation features need to be improved. The result explanation interface may include simulation examples and counter-simulation examples.

[0098] As an implementation, the simulation example may include a result explanation diagram, using a graphical tool to dynamically display each performance evaluation feature in the result explanation interface, and the user may operate (e.g., click, drag, etc.) on the result explanation interface to view the detailed information of each performance evaluation feature, and the user may also control the control corresponding to each performance evaluation feature, adjust the performance evaluation feature, and view the impact of the performance evaluation feature on the ESG rating result, and the user may also operate the local result explanation diagram in the result explanation interface to view the analysis of the ESG rating result generated according to each performance evaluation feature. The result explanation diagram may include a scatter plot, a decomposition diagram, a heat map, a bar chart, a line chart, etc.

[0099] As another implementation, the simulation example may also include a dynamic feature impact animation, which uses animation and transition effects to show the dynamic impact of changes in performance assessment features on ESG rating results, which can help users intuitively understand the changes in ESG rating results caused by changes in performance assessment features. Interactive controls can be set in the dynamic feature impact animation, and users can adjust the performance assessment features by controlling the interactive controls or in the interactive space. The dynamic feature impact animation can display the adjusted ESG rating results and the scores corresponding to each performance assessment feature in real time.

[0100] The ESG rating method provided by the embodiment of the present invention can enable users to clearly understand the influencing factors of ESG rating results through visualization technology and interactive functions.

[0101] In some embodiments, the electronic device can run an ESG rating system and a management system of the enterprise to be evaluated. The ESG rating system can execute the above-mentioned ESG rating method. After obtaining the ESG rating results of the enterprise to be evaluated, the ESG rating system can send the ESG rating results of the enterprise to be evaluated to the management system of the enterprise to be evaluated to provide management assistance information for the enterprise to be evaluated.

[0102] Managers of the companies to be evaluated can view the ESG rating results from the management system of the companies to be evaluated and adjust the management policies of the companies to be evaluated based on the ESG rating results.

[0103] Among them, the electronic equipment running the ESG rating system may be the same electronic equipment that runs the management system of the enterprise to be evaluated, and the ESG rating system may be integrated into the management system of the enterprise to be evaluated; the electronic equipment running the ESG rating system may or may not be the same electronic equipment that runs the management system of the enterprise to be evaluated.

[0104] In order to execute the corresponding steps in the above embodiments and various possible methods, a method for implementing an ESG rating device is provided below. Figure 5 , is a functional module diagram of an ESG rating device provided by an embodiment of the present invention. It should be noted that the basic principle and technical effects of the ESG rating device provided by this embodiment are the same as those of the above embodiments. For the sake of brief description, for parts not mentioned in this embodiment, please refer to the corresponding contents in the above embodiments. The ESG rating device 200 includes: an enterprise performance data acquisition module 210, an enterprise performance data processing module 220, a performance feature extraction module 230 and an ESG rating result acquisition module 240, wherein:

[0105] The enterprise performance data acquisition module 210 is used to collect a plurality of enterprise performance data from different data sources; the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance.

[0106] The enterprise performance data processing module 220 is used to perform standardization processing on a plurality of enterprise performance data to obtain preliminary enterprise performance data.

[0107] The performance feature extraction module 230 is used to extract features from the preliminary enterprise performance data to obtain performance features.

[0108] The ESG rating result acquisition module 240 is used to input the performance characteristics into the trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated; wherein the trained ESG rating model is trained based on the historical performance characteristics of multiple enterprises and the ESG rating results of each enterprise.

[0109] Furthermore, the enterprise performance data processing module 220 is specifically used to pre-process multiple enterprise performance data to obtain multiple pre-processed enterprise performance data; perform deep cleaning on the multiple pre-processed enterprise performance data; and merge the multiple pre-processed enterprise performance data after deep cleaning to obtain preliminary enterprise performance data.

[0110] Furthermore, the enterprise performance data processing module 220 is specifically used to perform preliminary cleaning on the plurality of enterprise performance data; convert the plurality of enterprise performance data after preliminary cleaning into a unified format to obtain a plurality of pre-processed enterprise performance data.

[0111] Furthermore, the performance characteristics include multiple performance evaluation characteristics, and the ESG rating result acquisition module 240 is specifically used to calculate the score corresponding to each performance evaluation characteristic; according to the preset weight and score corresponding to each performance evaluation characteristic, the ESG rating result of the enterprise to be evaluated is obtained.

[0112] Furthermore, the ESG rating result acquisition module 240 is specifically used to generate a result explanation interface based on the relationship between each performance evaluation feature and the ESG rating result of the enterprise to be evaluated; the result explanation interface is used to display the impact of each performance evaluation feature on the ESG rating result of the enterprise to be evaluated.

[0113] Furthermore, the ESG rating device 200 further includes: an ESG rating model training module, wherein:

[0114] The ESG rating model training module is used to obtain training samples and the annotation results corresponding to the training samples: the training samples are the historical performance characteristics of each enterprise, and the annotation results are the ESG rating results of each enterprise; the training samples are input into a pre-established machine learning model to obtain the predicted ESG rating results corresponding to the training samples; the parameters of the machine learning model are iteratively updated according to the predicted ESG rating results corresponding to the training samples and the annotation results to obtain a trained ESG rating model.

[0115] Furthermore, the ESG rating device 200 further includes: an ESG rating result sending module, wherein:

[0116] The ESG rating result sending module is used to send the ESG rating results of the enterprise to be evaluated to the management system of the enterprise to be evaluated, so as to provide management auxiliary information for the enterprise to be evaluated.

[0117] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0118] In several embodiments provided by the present invention, the coupling between modules may be electrical, mechanical or other forms of coupling.

[0119] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0120] Please refer to Figure 6 , is a block diagram of an electronic device 100 provided in an embodiment of the present invention. The electronic device 100 includes a memory 110, a processor 120, and a communication module 130. The memory 110, the processor 120, and the communication module 130 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0121] The memory 110 is used to store programs or data. The memory 110 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0122] The processor 120 is used to read / write data or programs stored in the memory and execute corresponding functions. For example, when the computer program stored in the memory 110 is executed by the processor 120, the ESG rating method disclosed in the above embodiments can be implemented.

[0123] The communication module 130 is used to establish a communication connection between the electronic device 100 and other electronic devices through a network, and to send and receive data through the network.

[0124] It should be understood that Figure 6 The structure shown is only a schematic diagram of the structure of the electronic device. The electronic device may also include Figure 6 More or fewer components may be shown. Figure 6 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0125] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program can be executed by a processor, the ESG rating method described in the above method embodiment is implemented.

[0126] The computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program codes that execute any of the method steps in the above method. These program codes can be read from or written to one or more computer program products. The program code can be compressed, for example, in an appropriate form.

[0127] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a module, a program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0128] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0129] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An ESG rating method, characterized in that: The method comprises: Collecting multiple enterprise performance data from different data sources; the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance; Standardizing the plurality of enterprise performance data to obtain preliminary enterprise performance data; Performing feature extraction on the preliminary enterprise performance data to obtain performance features; The performance characteristics are input into a trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated; wherein the trained ESG rating model is trained based on the historical performance characteristics of multiple enterprises and the ESG rating results of each of the enterprises.

2. The method according to claim 1, characterized in that The step of performing standardization on the plurality of enterprise performance data to obtain preliminary enterprise performance data includes: Preprocessing the plurality of enterprise performance data to obtain a plurality of preprocessed enterprise performance data; Deeply cleaning the plurality of pre-processed enterprise performance data; The multiple pre-processed enterprise performance data after deep cleaning are merged to obtain the preliminary enterprise performance data.

3. The method according to claim 2, characterized in that The preprocessing of the plurality of enterprise performance data to obtain a plurality of preprocessed enterprise performance data includes: Conduct preliminary cleaning of the performance data of multiple companies; The plurality of enterprise performance data after preliminary cleaning are converted into a unified format to obtain the plurality of pre-processed enterprise performance data.

4. The method according to claim 1, characterized in that: The performance features include a plurality of performance evaluation features, and the step of inputting the performance features into the trained ESG rating model to obtain the ESG rating result of the enterprise to be evaluated includes: Calculating a score corresponding to each of the performance assessment characteristics; According to the preset weight and score corresponding to each performance evaluation feature, the ESG rating result of the enterprise to be evaluated is obtained.

5. The method according to claim 4, characterized in that The method further comprises: A result explanation interface is generated according to the relationship between each of the performance evaluation characteristics and the ESG rating result of the enterprise to be evaluated; the result explanation interface is used to display the impact of each of the performance evaluation characteristics on the ESG rating result of the enterprise to be evaluated.

6. The method according to claim 1, characterized in that The trained ESG rating model is trained by the following steps: Obtain training samples and the annotation results corresponding to the training samples: The training samples are historical performance characteristics of each of the enterprises, and the annotation results are ESG rating results of each of the enterprises; Inputting the training sample into a pre-established machine learning model to obtain a predicted ESG rating result corresponding to the training sample; The parameters of the machine learning model are iteratively updated according to the predicted ESG rating results corresponding to the training samples and the annotation results to obtain a trained ESG rating model.

7. The method according to claim 1, characterized in that After obtaining the ESG rating result of the enterprise to be evaluated, the method further includes: The ESG rating result of the enterprise to be evaluated is sent to the management system of the enterprise to be evaluated to provide management auxiliary information for the enterprise to be evaluated.

8. An ESG rating device, characterized in that: The device comprises: An enterprise performance data acquisition module is used to collect a plurality of enterprise performance data from different data sources; the enterprise performance data is the performance data of the enterprise to be evaluated in terms of environment, society and governance; An enterprise performance data processing module, used for performing standardization processing on the plurality of enterprise performance data to obtain preliminary enterprise performance data; A performance feature extraction module, used to extract features from the preliminary enterprise performance data to obtain performance features; The ESG rating result acquisition module is used to input the performance characteristics into the trained ESG rating model to obtain the ESG rating results of the enterprise to be evaluated; wherein the trained ESG rating model is trained based on the historical performance characteristics of multiple enterprises and the ESG rating results of each of the enterprises.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the computer program executable by the processor is used to implement the ESG rating method described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the ESG rating method according to any one of claims 1 to 7 is implemented.

Images