Enterprise credit comprehensive evaluation report generation system and generation method

Through the enterprise credit comprehensive evaluation report generation system, combining historical and real-time data, multi-dimensional enterprise credit evaluation is carried out, the problem of credit rating lag is solved, the accuracy and reliability of evaluation reports are improved, and the rapid changes in market environment are adapted.

CN119027230APending Publication Date: 2024-11-26CHONGQING CREDIT INFORMATION CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411127750.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Existing corporate credit evaluation reports cannot promptly reflect current and future risk changes, resulting in lagging credit ratings and may lead to investors making wrong decisions.

Method used

A comprehensive enterprise credit evaluation report generation system was designed, including an evaluation report generation platform, user side and technical backend. By configuring the workbench, basic library, compile publisher, industry model library and storage, the configuration and compilation of industry models are realized. , combine historical data and real-time data to conduct multi-dimensional corporate credit assessment.

Benefits of technology

It improves the accuracy and reliability of corporate credit evaluation reports, ensures the real-time and reliability of evaluation results, and adapts to the personalized needs of different industries.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027230A_ABST
    Figure CN119027230A_ABST
Patent Text Reader

Abstract

The invention discloses an enterprise credit comprehensive evaluation report generation system and method, and the system comprises an evaluation report generation platform, and a user side and a technical background which are connected with the evaluation report generation platform through a network, and is characterized in that the evaluation report generation platform is provided with a configuration workbench; the configuration workbench is connected with a basic library and a compiling publisher, the compiling publisher is connected with an industry model library, the industry model library is connected with a memory, and the memory is connected with the user side. The method has the advantages that corresponding industry models are configured for different industry properties, historical data and real-time data are combined in the evaluation process, enterprise credit is comprehensively evaluated from multiple dimensions, and reliability and real-time performance of an evaluation report are effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise credit evaluation, and in particular to a system and method for generating an enterprise credit comprehensive evaluation report. Background Art

[0002] A corporate credit assessment report is a document prepared by information consulting agencies of financial institutions. Based on an enterprise's application and commission, the report comprehensively analyzes the enterprise's financial credit, economic benefits, management capabilities, and development prospects according to certain systematic indicators, and provides an impartial assessment of its creditworthiness. Its functions are manifested in the following ways: national regulatory departments need to use corporate credit assessments to adjust economic layout, conduct macroeconomic control, and optimize resource allocation; the financial system needs corporate credit to scientifically manage working capital and fully leverage the role of banks in the market economy; both parties in commodity transactions need to use corporate credit to avoid economic losses; and consumers need to use corporate credit to confidently purchase reliable goods.

[0003] Existing corporate credit rating reports are often based on past data and performance, failing to reflect current and future risk changes in a timely manner. Especially in a rapidly changing market environment, the lag in credit ratings may lead investors to make incorrect decisions. Summary of the Invention

[0004] The present invention provides a system and method for generating a comprehensive enterprise credit evaluation report, which can improve the reliability of the enterprise credit evaluation report.

[0005] To achieve the above objectives, the present invention provides a comprehensive enterprise credit evaluation report generation system, including an evaluation report generation platform, and a user terminal and a technical backend connected to the evaluation report generation platform via a network. The key features are: the evaluation report generation platform is equipped with a configuration workbench, which is connected to a basic library and a compiler / publisher; the compiler / publisher is connected to an industry model library, which is connected to a storage device, and the storage device is connected to the user terminal.

[0006] The configuration workbench is used to complete the configuration of the industry model;

[0007] The basic library stores various parameters, variables, and rules required for industry model configuration;

[0008] The compilation publisher is used to implement the compilation and publication of the industry model;

[0009] The industry model library stores various industry models. Users select an industry model that matches their own company as their industry user model, and enter relevant company information parameters to generate a corresponding comprehensive credit evaluation report.

[0010] Through the above design, different industry user models are configured for different industries to meet the personalized needs of credit evaluation content and indicators in each industry, effectively improving the accuracy and reliability of corporate credit evaluation reports.

[0011] As an example, the basic library is provided with a basic parameter library, a basic variable library, a data dictionary library, a basic rule library, a basic action library, a basic function library, a decision set, a decision table, a decision tree and a score card;

[0012] The basic parameter library includes but is not limited to result parameters, scoring parameters, quota parameters, interest rate parameters and weight parameters;

[0013] The basic variable library includes: three personal elements: applicant ID number, applicant name, and applicant mobile phone number; three basic corporate information elements: unified social credit code, company name, and name of the legal representative of the company;

[0014] The data dictionary library includes but is not limited to list types, rule file types, data suppliers, business scenario types, and monitoring indicators;

[0015] The basic rule base includes but is not limited to value range rules, start time rules, end time rules, null value processing rules, matching rules and inclusion rules;

[0016] The basic action library includes but is not limited to printing content to the console, variable assignment, function execution, and external data acquisition;

[0017] The basic function library includes date functions, mathematical functions and string functions;

[0018] The decision table includes a common decision table and a cross decision table;

[0019] The scoring cards include simple scoring cards and complex scoring cards.

[0020] A decision set is a set of predefined decision rules or conditions, typically used to guide the actions a system or program should take in specific situations. It's like a collection of decision elements, each with its own corresponding condition and action. For example, in software engineering, you can add attributes to a decision table, which are applied to each rule defined in the table.

[0021] A decision table is a tabular tool used to describe situations involving multiple combinations of decision conditions. It uses rows and columns to define problems and corresponding actions, enabling precise and concise descriptions of complex logical relationships. The advantage of a decision table is that it allows users to view business logic from different perspectives, providing answers to various decision-making questions. Furthermore, a decision table lists all possible scenarios and clearly indicates the corresponding handling methods, allowing users to easily identify which actions correspond to which situations, without having to consider the underlying logical relationships.

[0022] A decision tree is a graphical representation used to show the sequence of conditions and allowed actions. It is presented as a diagram, similar to a flow chart, consisting of internal nodes, branches, and leaf nodes. Each node represents a test attribute, each branch represents a possible value for that attribute, and the leaf nodes represent the final classification or decision label. The advantages of decision trees include being intuitive, easy to understand, and simple to maintain, making them particularly suitable for non-technical personnel in an organization.

[0023] Through the above design, a basic library is established, and then the corresponding industry model is configured according to the definition of the industry model, which effectively improves the efficiency and stability of model configuration.

[0024] Preferably, the industry model library contains m industry models, and each of the industry models is provided with a data collector, a data cleaner, a classification workbench, n classification indicator scoring workbenches, n classification aggregations, an overall aggregation, a reverse scorer, a rating workbench, and an evaluation workbench;

[0025] The data collector is connected to the data cleaner, the data cleaner is connected to the classification workbench, the classification workbench is connected to n classification indicator scoring workbenches, each of the classification indicator scoring workbenches is connected to a classification aggregation, all the classification aggregations are connected to the same overall aggregation, the overall aggregation is connected to the reverse scorer, the reverse scorer is connected to the rating workbench, and the rating workbench is connected to the evaluation workbench.

[0026] The data collector is used to collect enterprise-related information data.

[0027] The classification workbench is provided with classification rules, and the classification rules are for classifying data according to the scope of content covered by the data.

[0028] The classification indicator scoring workbench is used to score various indicators.

[0029] The rating workbench is provided with a rating library, which is used to rate the enterprise to be rated according to the overall rating result, including four grades: A, B, C, and D.

[0030] Among them, the grade A score is between 300 and 400, the grade B score is between 260 and 300, the grade C score is between 200 and 260, and the grade D score is between 0 and 200.

[0031] The evaluation workbench is used to perform text evaluation based on rating results, marking conditions and other contents.

[0032] The memory is used to store the generated enterprise credit comprehensive evaluation report and feed it back to the user.

[0033] Preferably, the data collector is provided with an interface collection unit and a database collection unit, and the data cleaner is provided with a marking unit, a type conversion unit and a cleaning unit.

[0034] The data collector collects the enterprise information parameters through a user interface, collects the external data through an external data interface, and collects the internal data through a database.

[0035] The data collector not only collects internal data, but also collects external data in real time through an external data interface, thus ensuring the real-time and validity of the data.

[0036] The marking unit designs special rules for certain special data in enterprise information data 'a' and marks them. After marking, special logic can be executed. For example, if the enterprise to be scored is included in the list of seriously dishonest entities, or is a dishonest person subject to enforcement, or its current business status is revoked or cancelled, then the "parameter_individual loan model level" is assigned the value D and the "parameter_individual loan model total score" is assigned the value 0. In subsequent processes, the scoring logic can be skipped.

[0037] The type conversion unit is used to convert the data type in the enterprise information data a, for example, converting text into numbers and converting time into current relative time.

[0038] The cleaning unit is configured with cleaning rules, including discarding, completing, not processing, and truth value conversion. Users can choose "filtering" or "replacing" as the cleaning method and set corresponding rules according to the fields to perform operations such as removing spaces and special characters.

[0039] As a preferred embodiment, the classification index scoring workbench is equipped with an yes / no calculator, a range calculator, and a text calculator;

[0040] Both the classification aggregation and the overall aggregation are equipped with an aggregator, which is used to aggregate and add up the scoring data or scoring results.

[0041] The expression of the classification aggregation is as follows:

[0042]

[0043] Among them, Y j represents the j-th category scoring result, k represents the number of scoring data, y ji Represents the i-th j-category scoring data;

[0044] The overall aggregation expression is as follows:

[0045]

[0046] Among them, Z represents the overall scoring result, and n represents the number of scoring results;

[0047] The reverse scoring device is used to perform reverse scoring calculation on the overall scoring result based on the enterprise information data a, and its expression is as follows:

[0048]

[0049] Wherein, H represents the reverse scoring result, W and U represent two inversely proportional scoring index data in the enterprise information data a, β represents the correlation coefficient of the impact score index data W and U.

[0050] The range calculator is used to score specific numerical ranges in the data. For example, if the operator's age is greater than or equal to 25 and less than 30, or greater than or equal to 50 and less than 55, then the "Parameter_Operator's Age Score" will be assigned a value of 18.

[0051] The calculator is used to score specific numerical values ​​in the data. For example, if the number of currently valid administrative licenses is empty, then the value of "Parameter_Currently Valid Administrative License Quantity Score" will be set to 0.

[0052] The text calculator is used to score specific text information in the data. For example, if the operator's educational background is "Master's degree", then the parameter _Operator's educational background score is assigned a value of 10.

[0053] The reverse scoring device is used to reversely score inversely proportional information data within the enterprise information data. For example, when an enterprise has been established for a short time but has a large registered capital, financing funds, or enterprise scale, the enterprise's establishment time is inversely proportional to the enterprise's registered capital, financing funds, or enterprise scale. The overall score result is multiplied by the ratio of the enterprise's establishment time to the enterprise's registered capital, financing funds, or enterprise scale to perform a reverse scoring calculation, and then multiplied by the correlation coefficient to accurately adjust the overall score result.

[0054] The correlation coefficient can be defined based on information such as shareholder information, investor information, endorser information, and guarantee information from the company's public information. For example, if the shareholder is a central state-owned enterprise, the correlation coefficient is set to 80; if the shareholder is a local state-owned enterprise, the correlation coefficient is set to 70; and if the shareholder is a large listed company, the correlation coefficient is set to 60. Further correlation coefficients can be set based on whether the shareholder, investor, endorser, or guarantor is a Fortune 500 company and its specific ranking.

[0055] The scoring operations for different types of data are achieved through range calculators, yes / no calculators, and text calculators, making the scoring results more comprehensive and reliable. The setting of the reverse scorer enables the evaluation report generation platform to not only base its scoring on the independent scores of each indicator, but also to make comprehensive judgments on various indicators, effectively improving the accuracy of the scoring results.

[0056] Preferably, the evaluation workbench is provided with an evaluation library, a marking processing library and a special processing library.

[0057] The evaluation workbench is used to evaluate the companies to be evaluated based on the rating results, where A is rated as excellent, B as good, C as average, and D as poor.

[0058] The special processing library is mainly used to impose further level restrictions on the already evaluated results, i.e., a veto. For example, if the limit on high-consumption items is not empty and is greater than 0, and the credit report evaluation model score is greater than 270, then: "Parameter_Credit Report Evaluation Model Rating" is assigned the value BBB.

[0059] The labeling processing library is used to perform special processing on already labeled data. For example, in step 3, after labeling the enterprise to be evaluated with serious violations as D, it can directly enter the evaluation step; or the labeled content can be evaluated in a targeted manner.

[0060] A method for generating a comprehensive enterprise credit evaluation report comprises the following steps:

[0061] Step 1: Based on the industry type of the company to be evaluated, the user selects the corresponding industry model from the industry model library as the industry user model, and enters the company information parameters of the company to be evaluated into the industry user model through the user interface.

[0062] Step 2: The data collector in the industry user model collects enterprise information data a associated with the enterprise to be evaluated and passes it to the data cleaner;

[0063] Step 3: The data cleaner cleans the enterprise information data a to obtain enterprise standard data b, and passes it to the classification workbench;

[0064] Step 4: The classification workbench performs a classification operation on the enterprise standard data b, dividing it into category A data X1, category B data X2, category C data X3, and category D data X4. Category A data X1 is then passed to the A classification indicator scoring workbench, category B data X2 is passed to the B classification indicator scoring workbench, category C data X3 is passed to the C classification indicator scoring workbench, and category D data X4 is passed to the D classification indicator scoring workbench.

[0065] Step 5: The A-category scoring workbench performs a credit score on A-category data X1 and passes the A-category score data y1 to the A-category aggregation; the B-category scoring workbench performs a credit score on B-category data X2 and passes the B-category score data y2 to the B-category aggregation; the C-category scoring workbench performs a credit score on C-category data X3 and passes the C-category score data y3 to the C-category aggregation; the D-category scoring workbench performs a credit score on D-category data X4 and passes the D-category score data y4 to the D-category aggregation.

[0066] Step 6: The A-category aggregation adds the A-category score data y1 together to obtain the A-category score result Y1, and then passes it to the overall aggregation; the B-category aggregation adds the B-category score data y2 together to obtain the B-category score result Y2, and then passes it to the overall aggregation; the C-category aggregation adds the C-category score data y3 together to obtain the C-category score result Y3, and then passes it to the overall aggregation; the D-category aggregation adds the D-category score data y4 together to obtain the D-category score result Y4, and then passes it to the overall aggregation.

[0067] Step 7: The overall aggregation performs overall aggregation on the A-class score result Y1, B-class score result Y2, C-class score result Y3 and D-class score result Y4 to obtain the overall score result Z, and passes it to the reverse scorer;

[0068] Step 8: The reverse scorer performs reverse score calculation on the overall score result Z to obtain the reverse score result H, and transmits it to the rating workbench;

[0069] Step 9: The rating workbench performs credit rating based on the reverse scoring result H, obtains rating result G, and transmits it to the evaluation workbench;

[0070] Step 10: The evaluation workbench generates a credit evaluation report based on the rating result G and stores it in the memory;

[0071] Step 11: The memory feeds back the credit evaluation report to the user terminal through the user interface.

[0072] Through the above design, multi-level indicators are used to conduct a comprehensive corporate credit evaluation of the evaluated enterprises from different dimensions, which effectively improves the reliability of the corporate credit evaluation report.

[0073] Preferably, in step 1, the industry model is generated by the following steps:

[0074] Step A1: The technical backend defines the industry model through the technical interface and builds the basic library;

[0075] Step A2: The configuration workbench obtains the corresponding parameter information from the basic library according to the nature of the industry model, configures the industry model, and passes the configured industry model to the compiler and publisher;

[0076] Step A3: The compiler publisher compiles and publishes the configured industry model to the industry model library.

[0077] Through the above design, different industry models are defined according to the different nature of the industry, and targeted industry model configuration is achieved, thereby improving the accuracy and reliability of the scoring, rating and evaluation operations of the corresponding industry enterprise information data by each industry model.

[0078] Preferably, in step 2, the enterprise information data a includes: enterprise information parameters entered by the user, internal data related to the enterprise to be evaluated stored in the internal database, and external data related to the enterprise to be evaluated obtained in real time through an external data interface.

[0079] The data collector collects the enterprise information parameters through a user interface, collects the external data through an external data interface, and collects the internal data through a database.

[0080] Through the above design, historical data is combined with real-time data to comprehensively evaluate the credit status of enterprises and ensure the reliability and real-time nature of the evaluation report.

[0081] Preferably, in step 4, the A-type data X1 is basic data on trustworthiness, the B-type data X2 is dynamic business data, the C-type data X3 is data on compliance with laws and regulations, and the D-type data X4 is credit record data.

[0082] The basic data on trustworthiness includes the operator's age, number of currently valid administrative licenses, education level, years of operation, number of individual businesses currently operated by the operator, and number of times the operator serves as the legal representative of an enterprise.

[0083] The business dynamics data include the number of valid trademarks, the number of copyrights of works and the number of patents, etc.

[0084] The law-abiding data include the results of “double random, one public” spot checks, the amount of administrative penalties, etc.

[0085] The credit record data includes the number of positive evaluation types for the performance of public resource transaction contracts, tax credit rating, etc.

[0086] Comprehensive corporate credit evaluation focuses on a holistic assessment of an enterprise, distinct from public and industry credit evaluations. It comprehensively integrates credit information generated throughout an enterprise's life cycle, encompassing three categories: public credit indicators, industry credit indicators, and operational credit indicators. Public credit indicators examine a company's past behavior and credit performance from the perspective of violations and breaches of trust. A company's lack of a record of violations does not necessarily mean it will not engage in similar behavior in the future. A record of violations merely provides a reference for a company's credit profile and cannot fully predict its future behavior. Industry credit indicators assess the overall credit risk of the industry within which it operates, revealing common and systemic risks within the industry. When overall industry risk is high, so too is the corresponding risk for the company. Therefore, industry credit evaluation can serve as an important supplement to assessing corporate credit. Operational credit indicators assess a company's operational capabilities, strength, stability, and governance by examining fundamental information such as its operating status, registered capital, years of operation, employee size, and equity structure. These indicators reflect a company's operating conditions and management skills, thus revealing its future ability to fulfill its obligations. Based on this, the comprehensive corporate credit evaluation assesses enterprises holistically from four dimensions: corporate integrity, legal compliance, credit record, and business performance. It scientifically predicts the probability of future risks for enterprises and effectively improves the reliability and accuracy of credit evaluation reports.

[0087] The beneficial effects of the present invention are as follows: corresponding industry models are configured for different industry properties, and historical data is combined with real-time data during the evaluation process to conduct a comprehensive evaluation of corporate credit from multiple dimensions, effectively improving the reliability and real-time nature of the evaluation report. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 It is a structural diagram of the present invention;

[0089] Figure 2 is a flow chart of the present invention;

[0090] Figure 3 Schematic diagram of the industry user model scoring process in the embodiment. DETAILED DESCRIPTION

[0091] The present invention will be further described in detail below with reference to the accompanying drawings and specific examples. The following examples or drawings are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0092] like Figure 1 、 Figure 2 As shown: A comprehensive enterprise credit evaluation report generation system includes an evaluation report generation platform 1, and a user terminal 2 and a technical backend 3 connected to the evaluation report generation platform 1 via a network. The evaluation report generation platform 1 is equipped with a configuration workbench 8, which is connected to a basic library 4 and a compiler and publisher 5. The compiler and publisher 5 is connected to an industry model library 6, which is connected to a storage device 7, and the storage device 7 is connected to the user terminal 2.

[0093] The configuration workbench 8 is used to complete the configuration of the industry model;

[0094] The basic library stores various parameters, variables, and rules required for industry model configuration;

[0095] The compilation publisher is used to implement the compilation and publication of the industry model;

[0096] The industry model library stores various industry models. Users select an industry model that matches their own company as their industry user model, and enter relevant company information parameters to generate a corresponding comprehensive credit evaluation report.

[0097] The basic library 4 is provided with a basic parameter library, a basic variable library, a data dictionary library, a basic rule library, a basic action library, a basic function library, a decision set, a decision table, a decision tree and a score card;

[0098] The basic parameter library includes but is not limited to result parameters, scoring parameters, quota parameters, interest rate parameters and weight parameters;

[0099] The basic variable library includes: three personal elements: applicant ID number, applicant name, and applicant mobile phone number; three basic corporate information elements: unified social credit code, company name, and name of the legal representative of the company;

[0100] The data dictionary library includes but is not limited to list types, rule file types, data suppliers, business scenario types, and monitoring indicators;

[0101] The basic rule base includes but is not limited to value range rules, start time rules, end time rules, null value processing rules, matching rules and inclusion rules;

[0102] The basic action library includes but is not limited to printing content to the console, variable assignment, function execution, and external data acquisition;

[0103] The basic function library includes date functions, mathematical functions and string functions;

[0104] The decision table includes a common decision table and a cross decision table;

[0105] The scoring cards include simple scoring cards and complex scoring cards.

[0106] A decision set is a set of predefined decision rules or conditions, typically used to guide the actions a system or program should take in specific situations. It's like a collection of decision elements, each with its own corresponding condition and action. For example, in software engineering, you can add attributes to a decision table, which are applied to each rule defined in the table.

[0107] A decision table is a tabular tool used to describe situations involving multiple combinations of decision conditions. It uses rows and columns to define problems and corresponding actions, enabling precise and concise descriptions of complex logical relationships. The advantage of a decision table is that it allows users to view business logic from different perspectives, providing answers to various decision-making questions. Furthermore, a decision table lists all possible scenarios and clearly indicates the corresponding handling methods, allowing users to easily identify which actions correspond to which situations, without having to consider the underlying logical relationships.

[0108] A decision tree is a graphical representation used to show the sequence of conditions and allowed actions. It is presented as a diagram, similar to a flow chart, consisting of internal nodes, branches, and leaf nodes. Each node represents a test attribute, each branch represents a possible value for that attribute, and the leaf nodes represent the final classification or decision label. The advantages of decision trees include being intuitive, easy to understand, and simple to maintain, making them particularly suitable for non-technical personnel in an organization.

[0109] The industry model library 6 contains m industry models. Each industry model is equipped with a data collector, a data cleaner, a classification workbench, a scoring workbench with n classification indicators, n classification aggregations, an overall aggregation, a reverse scorer, a rating workbench, and an evaluation workbench.

[0110] The data collector is connected to the data cleaner, the data cleaner is connected to the classification workbench, the classification workbench is connected to n classification indicator scoring workbenches, each of the classification indicator scoring workbenches is connected to a classification aggregation, all the classification aggregations are connected to the same overall aggregation, the overall aggregation is connected to the reverse scorer, the reverse scorer is connected to the rating workbench, and the rating workbench is connected to the evaluation workbench.

[0111] The data collector is used to collect enterprise-related information data.

[0112] The classification workbench is provided with classification rules, and the classification rules are for classifying data according to the scope of content covered by the data.

[0113] The classification indicator scoring workbench is used to score various indicators.

[0114] The rating workbench is provided with a rating library, which is used to rate the enterprise to be rated according to the overall rating result, including four grades: A, B, C, and D.

[0115] Among them, the grade A score is between 300 and 400, the grade B score is between 260 and 300, the grade C score is between 200 and 260, and the grade D score is between 0 and 200.

[0116] The evaluation workbench is used to perform text evaluation based on rating results, marking conditions and other contents.

[0117] The memory is used to store the generated enterprise credit comprehensive evaluation report and feed it back to the user.

[0118] The data collector is provided with an interface collection unit and a database collection unit, and the data cleaner is provided with a marking unit, a type conversion unit and a cleaning unit.

[0119] The data collector collects the enterprise information parameters through a user interface, collects the external data through an external data interface, and collects the internal data through a database.

[0120] The data collector not only collects internal data, but also collects external data in real time through an external data interface, thus ensuring the real-time and validity of the data.

[0121] The marking unit designs special rules for certain special data in enterprise information data 'a' and marks them. After marking, special logic can be executed. For example, if the enterprise to be scored is included in the list of seriously dishonest entities, or is a dishonest person subject to enforcement, or its current business status is revoked or cancelled, then the "parameter_individual loan model level" is assigned the value D and the "parameter_individual loan model total score" is assigned the value 0. In subsequent processes, the scoring logic can be skipped.

[0122] The type conversion unit is used to convert the data type in the enterprise information data a, for example, converting text into numbers and converting time into current relative time.

[0123] The cleaning unit is configured with cleaning rules, including discarding, completing, not processing, and truth value conversion. Users can choose "filtering" or "replacing" as the cleaning method and set corresponding rules according to the fields to perform operations such as removing spaces and special characters.

[0124] The classification index scoring workbench is provided with a yes / no calculator, a range calculator and a text calculator;

[0125] Both the classification aggregation and the overall aggregation are equipped with an aggregator, which is used to aggregate and add up the scoring data or scoring results.

[0126] The expression of the classification aggregation is as follows:

[0127]

[0128] Among them, Y j represents the j-th category scoring result, k represents the number of scoring data, y ji Represents the i-th j-category scoring data;

[0129] The overall aggregation expression is as follows:

[0130]

[0131] Among them, Z represents the overall scoring result, and n represents the number of scoring results;

[0132] The reverse scoring device is used to perform reverse scoring calculation on the overall scoring result based on the enterprise information data a, and its expression is as follows:

[0133]

[0134] Wherein, H represents the reverse scoring result, W and U represent two inversely proportional scoring index data in the enterprise information data a, β represents the correlation coefficient of the impact score index data W and U.

[0135] The range calculator is used to score specific numerical ranges in the data. For example, if the operator's age is greater than or equal to 25 and less than 30, or greater than or equal to 50 and less than 55, then the "Parameter_Operator's Age Score" will be assigned a value of 18.

[0136] The calculator is used to score specific numerical values ​​in the data. For example, if the number of currently valid administrative licenses is empty, then the value of "Parameter_Currently Valid Administrative License Quantity Score" will be set to 0.

[0137] The text calculator is used to score specific text information in the data. For example, if the operator's educational background is "Master's degree", then the parameter _Operator's educational background score is assigned a value of 10.

[0138] The reverse scoring device is used to reversely score inversely proportional information data within the enterprise information data. For example, when an enterprise has been established for a short time but has a large registered capital, financing funds, or enterprise scale, the enterprise's establishment time is inversely proportional to the enterprise's registered capital, financing funds, or enterprise scale. The overall score result is multiplied by the ratio of the enterprise's establishment time to the enterprise's registered capital, financing funds, or enterprise scale to perform a reverse scoring calculation, and then multiplied by the correlation coefficient to accurately adjust the overall score result.

[0139] W = the number of years the company has been established, U = the sum of the company's cumulative net financing amount and the current planned financing, in millions. For example, if the company has been established for 2 years and the sum of the cumulative financing amount and the current planned financing is 500 million, then If the company has already distributed dividends of 100 million, then

[0140] The correlation coefficient can be defined based on information such as shareholder information, investor information, endorser information, and guarantee information from the company's public information. For example, if the shareholder is a central state-owned enterprise, the correlation coefficient is set to 80; if the shareholder is a local state-owned enterprise, the correlation coefficient is set to 70; and if the shareholder is a large listed company, the correlation coefficient is set to 60. Further correlation coefficients can be set based on whether the shareholder, investor, endorser, or guarantor is a Fortune 500 company and its specific ranking.

[0141] The scoring operations for different types of data are achieved through range calculators, yes / no calculators, and text calculators, making the scoring results more comprehensive and reliable. The setting of the reverse scorer enables the evaluation report generation platform to not only base its scoring on the independent scores of each indicator, but also to make comprehensive judgments on various indicators, effectively improving the accuracy of the scoring results.

[0142] The evaluation workbench is provided with an evaluation library, a marking processing library and a special processing library.

[0143] The evaluation workbench is used to evaluate the companies to be evaluated based on the rating results, where A is rated as excellent, B as good, C as average, and D as poor.

[0144] The special processing library is mainly used to impose further level restrictions on the already evaluated results, i.e., a veto. For example, if the limit on high-consumption items is not empty and is greater than 0, and the credit report evaluation model score is greater than 270, then: "Parameter_Credit Report Evaluation Model Rating" is assigned the value BBB.

[0145] The labeling processing library is used to perform special processing on already labeled data. For example, in step 3, after labeling the enterprise to be evaluated with serious violations as D, it can directly enter the evaluation step; or the labeled content can be evaluated in a targeted manner.

[0146] like Figure 2 A method for generating a comprehensive enterprise credit evaluation report includes the following steps:

[0147] Step 1: User terminal 2 selects the corresponding industry model from the industry model library 6 as the industry user model according to the industry type of the company to be evaluated, and enters the company information parameters of the company to be evaluated into the industry user model through the user interface;

[0148] Step 2: The data collector in the industry user model collects enterprise information data a associated with the enterprise to be evaluated and passes it to the data cleaner;

[0149] Step 3: The data cleaner cleans the enterprise information data a to obtain enterprise standard data b, and passes it to the classification workbench;

[0150] Step 4: The classification workbench performs a classification operation on the enterprise standard data b, dividing it into category A data X1, category B data X2, category C data X3, and category D data X4. Category A data X1 is then passed to the A classification indicator scoring workbench, category B data X2 is passed to the B classification indicator scoring workbench, category C data X3 is passed to the C classification indicator scoring workbench, and category D data X4 is passed to the D classification indicator scoring workbench.

[0151] Step 5: The A-category scoring workbench performs a credit score on A-category data X1 and passes the A-category score data y1 to the A-category aggregation; the B-category scoring workbench performs a credit score on B-category data X2 and passes the B-category score data y2 to the B-category aggregation; the C-category scoring workbench performs a credit score on C-category data X3 and passes the C-category score data y3 to the C-category aggregation; the D-category scoring workbench performs a credit score on D-category data X4 and passes the D-category score data y4 to the D-category aggregation.

[0152] Step 6: The A-category aggregation adds the A-category score data y1 together to obtain the A-category score result Y1, and then passes it to the overall aggregation; the B-category aggregation adds the B-category score data y2 together to obtain the B-category score result Y2, and then passes it to the overall aggregation; the C-category aggregation adds the C-category score data y3 together to obtain the C-category score result Y3, and then passes it to the overall aggregation; the D-category aggregation adds the D-category score data y4 together to obtain the D-category score result Y4, and then passes it to the overall aggregation.

[0153] Step 7: The overall aggregation performs overall aggregation on the A-class score result Y1, B-class score result Y2, C-class score result Y3 and D-class score result Y4 to obtain the overall score result Z, and passes it to the reverse scorer;

[0154] Step 8: The reverse scorer performs reverse score calculation on the overall score result Z to obtain the reverse score result H, and transmits it to the rating workbench;

[0155] Step 9: The rating workbench performs credit rating based on the reverse scoring result H, obtains rating result G, and transmits it to the evaluation workbench;

[0156] Step 10: The evaluation workbench generates a credit evaluation report based on the rating result G and stores it in the memory;

[0157] Step 11: The memory feeds back the credit evaluation report to the user terminal 2 through the user interface.

[0158] In step 1, the industry model is generated by:

[0159] Step A1: The technical backend 3 defines the industry model through the technical interface and builds the basic library 4;

[0160] Step A2: The configuration workbench 8 obtains the corresponding parameter information from the basic library 4 according to the nature of the industry model, configures the industry model, and passes the configured industry model to the compiler and publisher 5;

[0161] Step A3: The compiler and publisher 5 compiles and publishes the configured industry model to the industry model library 6.

[0162] Different industry models are defined based on the nature of different industries, enabling targeted configuration of industry models and thereby improving the accuracy and reliability of scoring, rating, and evaluation of enterprise information data in corresponding industries.

[0163] In step 2, the enterprise information data a includes: enterprise information parameters entered by the user terminal 2, internal data related to the enterprise to be evaluated stored in the internal database, and external data related to the enterprise to be evaluated obtained in real time through the external data interface.

[0164] The data collector collects the enterprise information parameters through a user interface, collects the external data through an external data interface, and collects the internal data through a database.

[0165] In step 4, the A-type data X1 is basic credit data, the B-type data X2 is operational dynamic data, the C-type data X3 is law-abiding data, and the D-type data X4 is credit record data.

[0166] The basic data on trustworthiness includes the operator's age, number of currently valid administrative licenses, education level, years of operation, number of individual businesses currently operated by the operator, and number of times the operator serves as the legal representative of an enterprise.

[0167] The business dynamics data include the number of valid trademarks, the number of copyrights of works and the number of patents, etc.

[0168] The law-abiding data include the results of “double random, one public” spot checks, the amount of administrative penalties, etc.

[0169] The credit record data includes the number of positive evaluation types for the performance of public resource transaction contracts, tax credit rating, etc.

[0170] like Figure 3 As shown: The industry user model scoring process is as follows:

[0171] (1) Start: The starting point of the process.

[0172] Prejudgment

[0173] (2) Preliminary judgment: Make some preliminary judgments and verifications.

[0174] (3) Whether the pre-judgment is passed: Check whether the pre-judgment is passed and decide the next step of the process.

[0175] Scoring process

[0176] (4) Basic credit rating: Credit rating is based on some basic data.

[0177] Total score of basic trustworthiness: the cumulative total score of basic trustworthiness score, used for subsequent calculations.

[0178] (5) Business performance rating: Evaluate and score the individual's business performance.

[0179] Total business dynamics score: The cumulative total score of the business dynamics score, which is an important part of the model score.

[0180] (6) Law-abidingness score: Assess and score the individual’s compliance with laws and regulations.

[0181] Total compliance score: The cumulative total score of the compliance score directly affects the model score.

[0182] (7) Credit record scoring: Evaluate and score an individual’s credit record.

[0183] Total credit score: The cumulative total score of the credit score, which is an important part of the model score.

[0184] Model scoring and rating

[0185] (8) Model scoring: Aggregate all scores into one model score.

[0186] (9) Rating level: The final rating level is determined based on the model rating.

[0187] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A system for generating a comprehensive credit evaluation report for an enterprise, comprising an evaluation report generating platform (1), and a user terminal (2) and a technical background (3) connected to the evaluation report generating platform (1) via a network, characterized in that: The evaluation report generation platform (1) is provided with a configuration workbench (8), the configuration workbench (8) is connected to a basic library (4) and a compiler publisher (5), the compiler publisher (5) is connected to an industry model library (6), the industry model library (6) is connected to a storage (7), and the storage (7) is connected to the user end (2).

2. The enterprise credit comprehensive evaluation report generation system according to claim 1 is characterized by: The basic library (4) is provided with a basic parameter library, a basic variable library, a data dictionary library, a basic rule library, a basic action library, a basic function library, a decision set, a decision table, a decision tree and a score card; The basic parameter library includes but is not limited to result parameters, scoring parameters, quota parameters, interest rate parameters and weight parameters; The basic variable library includes: three personal elements: applicant ID number, applicant name, and applicant mobile phone number; three basic corporate information elements: unified social credit code, corporate name, and name of the legal representative of the company; The data dictionary library includes but is not limited to list types, rule file types, data suppliers, business scenario types, and monitoring indicators; The basic rule base includes but is not limited to value range rules, start time rules, end time rules, null value processing rules, matching rules and inclusion rules; The basic action library includes but is not limited to printing content to the console, variable assignment, function execution and external data acquisition; The basic function library includes date functions, mathematical functions and string functions; The decision table includes a common decision table and a cross decision table; The scorecards include simple scorecards and complex scorecards.

3. The system for generating a comprehensive credit evaluation report for an enterprise according to claim 1, characterized in that: The industry model library (6) contains m industry models, each of which is provided with a data collector, a data cleaner, a classification workbench, n classification index scoring workbenches, n classification aggregations, an overall aggregation, a reverse scoring device, a rating workbench and an evaluation workbench; The data collector is connected to the data cleaner, the data cleaner is connected to the classification workbench, the classification workbench is connected to n classification indicator scoring workbenches, each of the classification indicator scoring workbenches is connected to a classification aggregation, all the classification aggregations are connected to the same overall aggregation, the overall aggregation is connected to the reverse scorer, the reverse scorer is connected to the rating workbench, and the rating workbench is connected to the evaluation workbench.

4. The system for generating a comprehensive enterprise credit evaluation report according to claim 3 is characterized in that: The data collector is provided with an interface collection unit and a database collection unit, and the data cleaner is provided with a marking unit, a type conversion unit and a cleaning unit.

5. The system for generating a comprehensive enterprise credit evaluation report according to claim 3 is characterized in that: The classification index scoring workbench is provided with a yes or no calculator, a range calculator and a text calculator; An aggregator is provided in both the classification aggregation and the overall aggregation, and the aggregator is used to aggregate and add the scoring data or the scoring results; The expression of the classification aggregation is as follows: Among them, Y j represents the j-th category scoring result, k represents the number of scoring data, y ji Represents the i-th j-category scoring data; The overall aggregation expression is as follows: Among them, Z represents the overall scoring result, and n represents the number of scoring results; The reverse scoring device is used to perform reverse scoring calculation on the overall scoring result according to the enterprise information data a, and its expression is as follows: Wherein, H represents the reverse scoring result, W and U represent two inversely proportional scoring index data in the enterprise information data a, β represents the correlation coefficient of the impact score index data W and U.

6. The system for generating a comprehensive enterprise credit evaluation report according to claim 3 is characterized in that: The evaluation workbench is provided with an evaluation library, a marking processing library and a special processing library.

7. A method for generating a comprehensive enterprise credit evaluation report, characterized in that: The method comprises the following steps: Step 1: the user terminal (2) selects a corresponding industry model from the industry model library (6) as an industry user model according to the industry type of the enterprise to be evaluated, and inputs the enterprise information parameters of the enterprise to be evaluated into the industry user model through the user interface; Step 2: The data collector in the industry user model collects the enterprise information data a associated with the enterprise to be evaluated and passes it to the data cleaner; Step 3: The data cleaner cleans the enterprise information data a to obtain enterprise standard data b, and passes it to the classification workbench; Step 4: The classification workbench performs classification operation on the enterprise standard data b, and divides it into Class A data X1, Class B data X2, Class C data X3 and Class D data X4, and passes Class A data X1 to the Class A index scoring workbench, passes Class B data X2 to the Class B index scoring workbench, passes Class C data X3 to the Class C index scoring workbench, and passes Class D data X4 to the Class D index scoring workbench; Step 5: The A classification indicator scoring workbench performs a credit score on the A data X1, and passes the A scoring data y1 to the A classification aggregation; the B classification indicator scoring workbench performs a credit score on the B data X2, and passes the B scoring data y2 to the B classification aggregation; the C classification indicator scoring workbench performs a credit score on the C data X3, and passes the C scoring data y3 to the C classification aggregation; the D classification indicator scoring workbench performs a credit score on the D data X4, and passes the D scoring data y4 to the D classification aggregation; Step 6: The A classification aggregation performs aggregation and addition on the A-class scoring data y1 to obtain the A-class scoring result Y1, and passes it to the overall aggregation; the B classification aggregation performs aggregation and addition on the B-class scoring data y2 to obtain the B-class scoring result Y2, and passes it to the overall aggregation; the C classification aggregation performs aggregation and addition on the C-class scoring data y3 to obtain the C-class scoring result Y3, and passes it to the overall aggregation; the D classification aggregation performs aggregation and addition on the D-class scoring data y4 to obtain the D-class scoring result Y4, and passes it to the overall aggregation; Step 7: The overall aggregation performs overall aggregation on the A-category scoring result Y1, the B-category scoring result Y2, the C-category scoring result Y3 and the D-category scoring result Y4 to obtain an overall scoring result Z, and passes it to the reverse scorer; Step 8: The reverse scorer performs reverse scoring calculation on the overall scoring result Z to obtain a reverse scoring result H, and passes it to the rating workbench; Step 9: The rating workbench performs credit rating according to the reverse rating result H, obtains a rating result G, and transmits it to the evaluation workbench; Step 10: The evaluation workbench makes a credit evaluation report according to the rating result G and stores it in a memory; Step 11: The memory feeds back the credit evaluation report to the user terminal (2) through the user interface.

8. The method for generating a comprehensive enterprise credit evaluation report according to claim 7, characterized in that: In step 1, the industry model is generated by the following steps: Step A1: The technical background (3) defines the industry model through the technical interface and builds the basic library (4); Step A2: The configuration workbench (8) obtains corresponding parameter information from the basic library (4) according to the properties of the industry model, configures the industry model, and transmits the configured industry model to the compilation publisher (5); Step A3: The compilation publisher (5) compiles and publishes the configured industry model to the industry model library (6).

9. The method for generating a comprehensive enterprise credit evaluation report according to claim 7, characterized in that: In step 2, the enterprise information data a includes: enterprise information parameters entered by the user end (2), internal data associated with the enterprise to be evaluated stored in the internal database, and external data associated with the enterprise to be evaluated obtained in real time through the external data interface.

10. The method for generating a comprehensive enterprise credit evaluation report according to claim 7, characterized in that: In step 4, the Class A data X1 is basic trustworthiness data, the Class B data X2 is business dynamic data, the Class C data X3 is discipline and law compliance data, and the Class D data X4 is credit record data.

Citation Information

Patent Citations

  • Enterprise credit evaluation system and method

    CN105719073A

  • Credit rating method based on enterprise credit reporting service

    CN107240014A

  • Enterprise credit dynamic scoring system and method thereof

    CN107527212A

  • Enterprise financing credit evaluation system and method

    CN112801779A