Enterprise scoring method and system, computer and storage medium

Abstract

The invention relates to the technical field of enterprise data processing, and provides an enterprise scoring method and system, a computer and a storage medium, and the enterprise scoring method comprises the steps: obtaining a plurality of pieces of enterprise data, and dividing the enterprise data into a plurality of dimension data sets according to an evaluation dimension; cleaning the dimension data set; filling data, and calculating uncertainty; constructing a data conversion rule, and determining a scoring weight; obtaining an index data set, and associating the uncertainty to the index data set; and calculating the total score of the enterprise and the uncertainty of the total score. According to the method, enterprise data is preprocessed, multiple evaluation dimensions and corresponding indexes are divided, calculation of uncertainty is added in the data preprocessing process, an accurate enterprise score is obtained, and the uncertainty of the overall evaluation result is quantified according to the score weight.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise data processing, and particularly to an enterprise scoring method, system, computer and storage medium. Background Art

[0002] In today's business environment, the evaluation of enterprises is of crucial significance to investors, partners, financial institutions, government regulatory departments, etc. Traditional enterprise evaluation methods often rely on single-dimensional data or subjective judgments, and there are problems such as incomplete, inaccurate and objective evaluations. For example, it is difficult to comprehensively reflect the actual situations in many aspects such as the innovation ability, market competitiveness, and social responsibility fulfillment of an enterprise only based on financial statement data.

[0003] With the development of big data technology, more and more enterprise data has been collected and stored. Existing enterprise data analysis usually conducts multi-dimensional data integration and comprehensive evaluation through scientific and reasonable analysis methods.

[0004] However, there are many channels to obtain enterprise data, the data sources are complex, the data quality is not stable enough, and it is difficult to process huge enterprise data. Even though the evaluation methods in the existing technology include data cleaning steps, the data cleaning lacks pertinence, and both the data cleaning and the enterprise evaluation results have uncertainties, and the uncertainties are difficult to intuitively quantify, making it difficult to judge the accuracy and reliability of the evaluation results. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an enterprise scoring method, system, computer and storage medium. The present invention solves the technical problems in the existing technology that it is difficult to process a large amount of complex enterprise data and lacks the evaluation of the accuracy of the evaluation results by means of a preprocessing step of cleaning data, dividing multiple evaluation dimensions and corresponding indicators, and adding the calculation of uncertainty during the process of preprocessing data, and quantifying the uncertainty of the overall evaluation result according to the scoring weight.

[0006] In a first aspect, an embodiment of the present application provides an enterprise scoring method, including the following steps:

[0007] Obtain a number of enterprise data of the enterprise to be evaluated, and divide the number of the enterprise data into a number of dimension data sets according to a number of evaluation dimensions;

[0008] Clean the dimension data sets to form a first preprocessed data set, and mark a number of missing data from the first preprocessed data set;

[0009] Based on the first preprocessed dataset and the several pieces of missing data, obtain several pieces of filled data, calculate the uncertainty, and fill the several pieces of filled data into the first preprocessed dataset to form a second preprocessed dataset;

[0010] Obtain several historical enterprise data and expert evaluation reports, establish several evaluation indicators according to each of the evaluation dimensions, construct a data conversion rule based on the several evaluation indicators and the several historical enterprise data, and establish a scoring weight based on the several evaluation dimensions and the expert evaluation report;

[0011] Based on the second preprocessed dataset and the data conversion rule, obtain an index dataset, and associate the uncertainty with the index dataset;

[0012] Calculate the total enterprise score and the uncertainty of the total score of the enterprise to be evaluated according to the several index datasets, the several uncertainties and the scoring weight.

[0013] Furthermore, the step of cleaning the dimension dataset to form a first preprocessed dataset and marking several pieces of missing data from the first preprocessed dataset includes:

[0014] Remove several pieces of incorrect data and several pieces of duplicate data from the dimension dataset to form a first preprocessed dataset;

[0015] Extract several content items and time items from the dimension dataset to establish a dimension data table, fill all the enterprise data in the first preprocessed dataset into the dimension data table to form a preprocessed data table, and mark several pieces of missing data according to the preprocessed data table.

[0016] Even further, the step of based on the first preprocessed dataset and the several pieces of missing data, obtaining several pieces of filled data, and calculating the uncertainty includes:

[0017] Train a BP neural network model based on the first preprocessed dataset to estimate several pieces of filled data;

[0018] Calculate several data means based on the preprocessed data table, and calculate the uncertainty based on the several data means, the total number of data in the preprocessed data table, the several pieces of filled data and the total number of the filled data.

[0019] Even further, the scoring weight includes a first-level sub-weight and a second-level sub-weight, and the step of establishing the scoring weight based on the several evaluation dimensions and the expert evaluation report includes:

[0020] Extract the historical total score from the expert evaluation report, establish a number of historical sub-scores corresponding one-to-one to a number of the evaluation dimensions based on the historical total score and the number of the evaluation dimensions, and establish the first-level sub-weights according to the historical total score and the number of the historical sub-scores;

[0021] Normalize a number of the historical enterprise data to form a normalized data set;

[0022] Based on the normalized data set and the data conversion rules, obtain a historical indicator data set, and establish the second-level sub-weights according to the historical indicator data set and a number of the historical sub-scores.

[0023] Furthermore, the step of calculating the enterprise total score and the total score uncertainty of the enterprise to be evaluated according to a number of the indicator data sets, a number of the uncertainties and the scoring weights includes:

[0024] Based on the second-level sub-weights and a number of the indicator data sets, obtain a number of enterprise sub-scores;

[0025] Based on the first-level sub-weights and a number of the enterprise sub-scores, obtain the enterprise total score;

[0026] Based on the first-level sub-weights and a number of the uncertainties, obtain the total score uncertainty.

[0027] Still further, after the step of calculating the enterprise total score and the total score uncertainty of the enterprise to be evaluated according to a number of the indicator data sets, a number of the uncertainties and the scoring weights, it further includes:

[0028] Generate an evaluation report based on the enterprise total score, the total score uncertainty and a number of the evaluation dimensions, and associate a number of the enterprise sub-scores to a number of the evaluation dimensions;

[0029] Based on the total score uncertainty and a preset uncertainty threshold, determine whether expert evaluation is required, generate an evaluation suggestion, and associate the evaluation suggestion to the evaluation report.

[0030] In a second aspect, an embodiment of the present application provides an enterprise scoring system, which is applied to the enterprise scoring method in the above technical solution, and the system includes:

[0031] An acquisition module, configured to acquire a number of enterprise data of an enterprise to be evaluated, and divide the number of the enterprise data into a number of dimension data sets according to a number of evaluation dimensions;

[0032] A cleaning module, configured to clean the dimension data sets to form a first preprocessed data set, and mark a number of missing data from the first preprocessed data set;

[0033] A filling module, configured to obtain a plurality of filled data based on the first preprocessed data set and the plurality of missing data, calculate the uncertainty, and fill the plurality of filled data into the first preprocessed data set to form a second preprocessed data set;

[0034] An establishing module, configured to obtain a plurality of historical enterprise data and expert evaluation reports, establish a plurality of evaluation indicators according to each of the evaluation dimensions, construct a data conversion rule based on the plurality of evaluation indicators and the plurality of historical enterprise data, and establish a scoring weight based on the plurality of evaluation dimensions and the expert evaluation report;

[0035] A conversion module, configured to obtain an index data set based on the second preprocessed data set and the data conversion rule, and associate the uncertainty with the index data set;

[0036] A scoring module, configured to calculate the total enterprise score and the total score uncertainty of the enterprise to be evaluated according to the plurality of index data sets, the plurality of uncertainties, and the scoring weight.

[0037] In a third aspect, an embodiment of the present application provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the enterprise scoring method described in the first aspect above is implemented.

[0038] In a fourth aspect, an embodiment of the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the enterprise scoring method described in the first aspect above is implemented.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows: before data cleaning, the enterprise data is divided in multiple dimensions. The evaluation dimensions may include financial dimension, market dimension, operation dimension, etc. After dividing several pieces of the enterprise data into several dimension data sets, targeted cleaning is performed, avoiding the problem that the data lacks differences in different dimensions due to unified cleaning of data in each dimension. After dividing the dimensions, the determination of incorrect and duplicate data will be more meticulous and conform to the situation of that dimension. The preprocessed data in different dimensions is trained and the missing data is predicted through the convolutional neural network technology. The predicted data is more accurate, avoiding the influence of data in different dimensions and being more in line with the characteristics of the dimension data set, and supplementing the first preprocessing data set. Since the filled data is not real data, the calculation of the uncertainty is added during the supplementing process. According to the expert review opinions, several pieces of the historical enterprise data are analyzed to establish several evaluation indicators affecting the evaluation dimension, and the second preprocessing data set is normalized and transformed to form the index data set that can be used for scoring. According to the scoring weights, a quantitative evaluation of the enterprise, that is, the total enterprise score, is finally formed. At the same time, enterprise sub-scores are also formed in multiple evaluation dimensions, and the evaluation results are clearer and more meticulous, meeting the different evaluation needs of evaluators and having strong applicability; the uncertainty is transmitted during the two-level scoring process to form the total score uncertainty of the total enterprise score, which can be used to reference the accuracy of the evaluation result and quantify the reliability of the evaluation result. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flowchart of the enterprise scoring method in the first embodiment of the present invention;

[0041] Figure 2 It is a structural block diagram of the enterprise scoring system in the second embodiment of the present invention;

[0042] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. SPECIFIC EMBODIMENTS

[0043] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0044] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there can also be an intermediate element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this invention belongs. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0046] Please refer to Figure 1 , the enterprise scoring method in the embodiment of the present invention includes the following steps:

[0047] Step S10: Obtain a number of enterprise data of the enterprise to be evaluated, and divide the number of said enterprise data into a number of dimension data sets according to a number of evaluation dimensions;

[0048] The enterprise data comes from multiple data sources, and the data sources include but are not limited to enterprise financial databases, enterprise operation management systems, market monitoring agency data platforms, regulatory department databases, etc. It can be understood that a large amount of data collected includes various types of data such as financial data, operation data, and market data. The evaluation dimensions are set according to various categories, and the number of said enterprise data is divided.

[0049] Step S20: Clean the dimension data set to form a first preprocessed data set, and mark a number of missing data from the first preprocessed data set;

[0050] Data cleaning usually clears duplicate and redundant data, completes missing data, corrects or removes incorrect data, which is beneficial to improving data quality. The processing scope of data cleaning includes missing value processing, outlier processing, duplicate data removal, data consistency processing, data normalization, data discretization, data type conversion, etc. Preferably, in the embodiment of the present application, incorrect and duplicate data are preferentially processed.

[0051] Specifically, the step S20 includes:

[0052] S210: Remove a number of incorrect data and a number of duplicate data from the dimension data set to form a first preprocessed data set;

[0053] S220: Extract several content items and time items from the dimension dataset to establish a dimension data table, and fill all the enterprise data in the first preprocessing dataset into the dimension data table to form a preprocessing data table, and mark several missing data according to the preprocessing data table.

[0054] Preferably, when evaluating the enterprise to be evaluated, it is necessary to examine the operation and development process of the enterprise. The enterprise data is usually based on time and can reflect the business conditions of the enterprise within time units such as monthly and quarterly. Based on time, organize the data reflecting the same content to establish the preprocessing data table, which is beneficial to finding out the missing data.

[0055] Step S30: Based on the first preprocessing dataset and several missing data, obtain several filling data, calculate the uncertainty, and fill the several filling data into the first preprocessing dataset to form a second preprocessing dataset;

[0056] Preferably, in traditional data cleaning methods, the processing of missing values usually adopts relatively simple mathematical methods such as mean filling, median filling, and mode filling. Even if a method with better fitting performance is used, it is impossible to avoid the problem that the filling data is different from the real data. Therefore, the calculation of the uncertainty is introduced, which is beneficial to quantifying the accuracy of the evaluation result. It can be understood that after setting several evaluation dimensions and dividing the data of different dimensions, fill in a targeted manner. The filling data is more in line with the characteristics of the corresponding dimension, and the uncertainty has a corresponding relationship with the evaluation dimension. When evaluating different dimensions of the enterprise to be evaluated subsequently, several uncertainties can provide more detailed references.

[0057] Specifically, the step S30 includes:

[0058] S310: Based on the first preprocessing dataset, train a BP neural network model to estimate several filling data;

[0059] It can be understood that taking the first preprocessing dataset as the training set, through the technology of deep learning, the filling data can fit the real data of the enterprise to be evaluated as much as possible. The BP neural network has strong non-linear fitting ability.

[0060] S320: Based on the preprocessing data table, calculate several data means, and calculate the uncertainty based on several data means, the total number of data in the preprocessing data table, several filling data, and the total number of filling data.

[0061] Preferably, the data can be grouped according to the content items and the data mean can be calculated. Understandably, the larger the total number of the filled data, the less real data there is in the data used for evaluation, and the higher the uncertainty of the evaluation result. If the filled data has a greater gap from the data mean, the possibility of data underfitting is higher.

[0062] Step S40: Obtain a number of historical enterprise data and expert evaluation reports, establish a number of evaluation indicators according to each of the evaluation dimensions, construct a data conversion rule based on the number of the evaluation indicators and the number of the historical enterprise data, and establish a scoring weight based on the number of the evaluation dimensions and the expert evaluation report;

[0063] Preferably, the evaluation of each evaluation dimension of an enterprise is affected by multiple factors. For example, for the financial dimension, the profitability, debt repayment ability, and growth ability of the enterprise all affect the financial evaluation of the enterprise, and can thus be used as a number of the evaluation indicators under the financial dimension. There are cases where one piece of enterprise financial data can reflect multiple evaluation indicators, and multiple pieces of data can reflect one evaluation indicator. Therefore, it is necessary to sort out the conversion rule from data to the evaluation indicators based on the existing expert evaluation report. Different evaluation indicators have different influences on the same evaluation dimension, and different evaluation dimensions have different influences on the overall evaluation of the enterprise to be evaluated. Therefore, it is necessary to obtain the scoring weight, that is, to quantify the analysis path from the evaluation indicators to the final evaluation result.

[0064] Specifically, the scoring weight includes a first-level sub-weight and a second-level sub-weight, and the step S40 includes:

[0065] S410: Extract the historical total score from the expert evaluation report, establish a number of historical sub-scores corresponding one-to-one to the number of the evaluation dimensions based on the historical total score and the number of the evaluation dimensions, and establish the first-level sub-weight according to the historical total score and the number of the historical sub-scores;

[0066] The first-level sub-weight can reflect the relationship between the scores on a number of the evaluation dimensions and the total score of the enterprise.

[0067] S420: Normalize the number of the historical enterprise data to form a normalized data set;

[0068] Understandably, normalizing the number of the historical enterprise data converts data in different formats and magnitudes into a unified standard format for subsequent processing and analysis.

[0069] S430: Obtain a historical index dataset based on the normalized dataset and the data transformation rule, and determine the secondary sub-weights according to the historical index dataset and several of the historical sub-ratings.

[0070] After processing the normalized dataset into the historical index dataset, a set of data that can intuitively reflect the level of the evaluation index can be obtained through mathematical methods. By using the historical sub-ratings and the historical index dataset, the rule from the index dataset to the sub-ratings is inversely deduced, that is, the secondary sub-weights. When experts examine the influence of different indicators and dimensions on the enterprise rating, the analytic hierarchy process can be used, and a judgment matrix can be established to assist in the analysis.

[0071] Step S50: Obtain an index dataset based on the second preprocessed dataset and the data transformation rule, and associate the uncertainty with the index dataset;

[0072] Several of the evaluation dimensions correspond to several dimension datasets. After cleaning and preprocessing, several of the second preprocessed datasets still correspond to several of the evaluation dimensions. After normalization processing and data transformation, several of the index datasets correspond to several of the evaluation dimensions. The uncertainties brought about during the preprocessing process also correspond to different evaluation dimensions respectively. Using several of the uncertainties is beneficial for quantifying the uncertainties of the data after preprocessing under multiple dimensions.

[0073] Step S60: Calculate the total enterprise rating and the uncertainty of the total rating of the enterprise to be evaluated according to several of the index datasets, several of the uncertainties, and the rating weights.

[0074] When conducting multi-level enterprise evaluations based on multiple evaluation dimensions, the uncertainty also undergoes transmission. After intuitively numerically quantifying the evaluation of the enterprise to be evaluated through the rating weights, the transmission results of the uncertainty are quantified using the same rule. The uncertainty of the total rating generated after quantification can be provided to the evaluators for reference.

[0075] Specifically, the step S60 includes:

[0076] S610: Obtain several enterprise sub-ratings based on the secondary sub-weights and several of the index datasets;

[0077] S620: Obtain the total enterprise rating based on the primary sub-weights and several of the enterprise sub-ratings;

[0078] S630: Obtain the uncertainty of the total rating based on the primary sub-weights and several of the uncertainties.

[0079] Understandably, in S610 to S620, not only the total enterprise score is obtained, but also the sub-enterprise scores under different dimensions can be obtained, making the evaluation of the enterprise more detailed and accurate. The scoring process is carried out in levels, and at the same time, the uncertainties brought by data cleaning and filling are also transmitted in levels.

[0080] After the step S60, it further includes:

[0081] S640: Generate an evaluation report based on the total enterprise score, the uncertainty of the total score, and several of the evaluation dimensions, and associate several of the sub-enterprise scores with several of the evaluation dimensions;

[0082] S650: Determine whether expert evaluation is required based on the uncertainty of the total score and a preset uncertainty threshold, generate an evaluation suggestion, and associate the evaluation suggestion with the evaluation report.

[0083] Understandably, the evaluation report not only includes multiple scores, but also includes reference opinions on whether each score is accurate enough. Several of the uncertainties correspond one-to-one with several of the evaluation dimensions. If the uncertainty is high, that is, the uncertainty of the total score exceeds the uncertainty threshold, a suggestion for manual evaluation is given to avoid misjudgment by the report reader.

[0084] Please refer to Figure 2 , the second embodiment of the present invention provides an enterprise scoring system, which is applied to the enterprise scoring method in the above embodiment, and the parts that have been described will not be repeated. As used below, terms such as "module", "unit", "sub-unit", etc. can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0085] The system includes:

[0086] An acquisition module 10, configured to acquire several enterprise data of an enterprise to be evaluated, and divide the several enterprise data into several dimension data sets according to several evaluation dimensions;

[0087] A cleaning module 20, configured to clean the dimension data sets to form a first preprocessed data set, and mark several missing data from the first preprocessed data set;

[0088] The cleaning module 20 includes:

[0089] A first unit, configured to remove several error data and several duplicate data from the dimension data sets to form a first preprocessed data set;

[0090] A second unit is used to extract a number of content items and time items from the dimension dataset to establish a dimension data table, fill all the enterprise data in the first preprocessed dataset into the dimension data table to form a preprocessed data table, and mark a number of missing data according to the preprocessed data table.

[0091] A filling module 30 is used to obtain a number of filling data based on the first preprocessed dataset and a number of the missing data, calculate the uncertainty, and fill the number of filling data into the first preprocessed dataset to form a second preprocessed dataset;

[0092] The filling module 30 includes:

[0093] A third unit is used to train a BP neural network model based on the first preprocessed dataset to estimate a number of filling data;

[0094] A fourth unit is used to calculate a number of data means based on the preprocessed data table, and calculate the uncertainty based on the number of data means, the total number of data in the preprocessed data table, the number of filling data, and the total number of the filling data.

[0095] An establishing module 40 is used to obtain a number of historical enterprise data and expert evaluation reports, establish a number of evaluation indicators according to each evaluation dimension, construct a data conversion rule based on the number of evaluation indicators and the number of historical enterprise data, and establish a scoring weight based on the number of evaluation dimensions and the expert evaluation report;

[0096] The establishing module 40 includes:

[0097] A fifth unit is used to extract the historical total score from the expert evaluation report, establish a number of historical sub-scores corresponding to the number of evaluation dimensions one by one based on the historical total score and the number of evaluation dimensions, and establish the first-level sub-weight according to the historical total score and the number of historical sub-scores;

[0098] A sixth unit is used to perform normalization processing on the number of historical enterprise data to form a normalized dataset;

[0099] A seventh unit is used to obtain a historical index dataset based on the normalized dataset and the data conversion rule, and establish the second-level sub-weight according to the historical index dataset and the number of historical sub-scores.

[0100] A conversion module 50 is used to obtain an index dataset based on the second preprocessed dataset and the data conversion rule, and associate the uncertainty with the index dataset;

[0101] A scoring module 60, configured to calculate the total enterprise score and the uncertainty of the total score of the enterprise to be evaluated according to a plurality of the index data sets, a plurality of the uncertainties, and the scoring weights.

[0102] The scoring module 60 includes:

[0103] An eighth unit, configured to obtain a plurality of enterprise sub-scores based on the secondary sub-weights and a plurality of the index data sets;

[0104] A ninth unit, configured to obtain the total enterprise score based on the primary sub-weights and a plurality of the enterprise sub-scores;

[0105] A tenth unit, configured to obtain the uncertainty of the total score based on the primary sub-weights and a plurality of the uncertainties;

[0106] An eleventh unit, configured to generate an evaluation report based on the total enterprise score, the uncertainty of the total score, and a plurality of the evaluation dimensions, and associate a plurality of the enterprise sub-scores with a plurality of the evaluation dimensions;

[0107] A twelfth unit, configured to determine whether expert evaluation is required based on the uncertainty of the total score and a preset uncertainty threshold, generate an evaluation suggestion, and associate the evaluation suggestion with the evaluation report.

[0108] The present invention further provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the enterprise scoring method described in the above technical solution is implemented.

[0109] The present invention further provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the enterprise scoring method described in the above technical solution is implemented.

[0110] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0111] The above-described embodiments merely represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention shall be subject to the appended claims.

Claims

1. A method for enterprise scoring, characterized in that: The steps include: Acquire a plurality of enterprise data of the enterprise to be evaluated, and divide the plurality of enterprise data into a plurality of dimensional data sets according to a plurality of evaluation dimensions; Cleaning the dimensional data set to form a first preprocessed data set, and marking a number of missing data from the first preprocessed data set; Based on the first preprocessed data set and the plurality of missing data, a plurality of filling data are obtained, and uncertainty is calculated, and the plurality of filling data are filled into the first preprocessed data set to form a second preprocessed data set; Acquire a number of historical enterprise data and expert evaluation reports, establish a number of evaluation indicators according to each of the evaluation dimensions, construct data conversion rules based on the number of evaluation indicators and the number of historical enterprise data, and establish scoring weights based on the number of evaluation dimensions and the expert evaluation reports; Based on the second preprocessed data set and the data conversion rule, obtaining an indicator data set, and associating the uncertainty to the indicator data set; According to the plurality of indicator data sets, the plurality of uncertainties and the scoring weights, the total enterprise score and the total score uncertainty of the enterprise to be evaluated are calculated.

2. The enterprise scoring method according to claim 1, characterized in that: The step of cleaning the dimensional data set to form a first preprocessed data set and marking a number of missing data from the first preprocessed data set includes: Removing a number of erroneous data and a number of duplicate data in the dimensional data set to form a first preprocessed data set; A number of content items and time items are extracted from the dimensional data set to establish a dimensional data table, and all the enterprise data in the first preprocessed data set are filled into the dimensional data table to form a preprocessed data table, and a number of missing data are marked according to the preprocessed data table.

3. The enterprise scoring method according to claim 2, characterized in that: The step of obtaining a plurality of filling data based on the first preprocessed data set and the plurality of missing data and calculating the uncertainty comprises: Based on the first preprocessed data set, training a BP neural network model to estimate a number of filling data; Based on the preprocessed data table, a number of data means are calculated, and based on the number of data means, the total number of data in the preprocessed data table, a number of the padded data and the total number of the padded data, the uncertainty is calculated.

4. The enterprise scoring method according to claim 1, characterized in that: The scoring weight includes a first-level sub-weight and a second-level sub-weight. Based on the evaluation dimensions and the expert evaluation report, the steps of establishing the scoring weight include: Extracting a historical total score from the expert evaluation report, establishing a number of historical sub-scores corresponding to the number of evaluation dimensions based on the historical total score and the number of evaluation dimensions, and establishing the first-level sub-weight according to the historical total score and the number of historical sub-scores; Normalizing the historical enterprise data to form a normalized data set; Based on the normalized data set and the data conversion rule, a historical indicator data set is obtained, and the secondary sub-weight is established according to the historical indicator data set and a plurality of the historical sub-scores.

5. The enterprise scoring method according to claim 4, characterized in that: The step of calculating the total enterprise score and the uncertainty of the total score of the enterprise to be evaluated according to the plurality of indicator data sets, the plurality of uncertainties and the scoring weights comprises: Based on the secondary sub-weights and the plurality of indicator data sets, a plurality of enterprise sub-scores are obtained; Based on the first-level sub-weights and the plurality of enterprise sub-scores, an overall enterprise score is obtained; Based on the first-level sub-weights and a number of the uncertainties, an overall score uncertainty is derived.

6. The enterprise scoring method according to claim 5, characterized in that: After the step of calculating the total enterprise score and the total score uncertainty of the enterprise to be evaluated according to the plurality of indicator data sets, the plurality of uncertainties and the score weights, the method further includes: Generate an evaluation report based on the total enterprise score, the uncertainty of the total score and the plurality of evaluation dimensions, and associate the plurality of enterprise sub-scores with the plurality of evaluation dimensions; Based on the total score uncertainty and a preset uncertainty threshold, it is determined whether expert evaluation is required, and an evaluation suggestion is generated, and the evaluation suggestion is associated with the evaluation report.

7. An enterprise scoring system, applied to the enterprise scoring method according to any one of claims 1 to 6, characterized in that: The system comprises: An acquisition module, used for acquiring a plurality of enterprise data of the enterprise to be evaluated, and dividing the plurality of enterprise data into a plurality of dimensional data sets according to a plurality of evaluation dimensions; a cleaning module, configured to clean the dimensional data set to form a first preprocessed data set, and mark a number of missing data from the first preprocessed data set; a filling module, configured to obtain a plurality of filling data based on the first preprocessed data set and the plurality of missing data, calculate uncertainty, and fill the plurality of filling data into the first preprocessed data set to form a second preprocessed data set; An establishment module is used to obtain a number of historical enterprise data and expert evaluation reports, establish a number of evaluation indicators according to each of the evaluation dimensions, construct data conversion rules based on the several evaluation indicators and the several historical enterprise data, and establish scoring weights based on the several evaluation dimensions and the expert evaluation reports; a conversion module, configured to obtain an indicator data set based on the second preprocessed data set and the data conversion rule, and associate the uncertainty with the indicator data set; The scoring module is used to calculate the total enterprise score and the total score uncertainty of the enterprise to be evaluated based on the plurality of indicator data sets, the plurality of uncertainties and the scoring weights.

8. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the enterprise scoring method according to any one of claims 1 to 6 is implemented.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the enterprise scoring method according to any one of claims 1 to 6 is implemented.

Images