Financial rating data optimization method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a financial rating data optimization method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: executing preprocessing operation on financial original data to obtain a standardized clean financial data set; performing rule verification on the clean financial data set, and outputting rule verification abnormal items and verification result data; based on the financial statement data in a preset time period and the calculated financial indexes, through the index group decision matrix, calculating to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level; calculating a financial health degree score and a financial health degree grade based on the calculated financial indexes and financial health degree model parameters; based on the financial anti-fraud risk level and the financial health degree level, constructing a comprehensive risk decision matrix; and generating a financial discrimination comprehensive risk level according to the comprehensive risk decision matrix. By adopting the method, the reliability and timeliness of credit quantitative rating can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of financial technology, in particular to a financial rating data optimization method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] In the process of credit quantitative rating of customers by financial institutions, quantitative rating is usually calculated based on the financial data of customers.

[0003] However, there are many problems in the current financial statements. First, because the data of some customers' online statements and offline statements are inconsistent, the rating results calculated based on the wrong data deviate from the truth; second, because some customers do not update the data used for rating in time, the rating results are out of date and invalid; third, because the income distribution of some group customers is randomly chaotic, the accuracy of the overall rating of group customers is affected; fourth, because the financial data of some customers do not conform to the business experience attempt, lack of business logic support, and thus the reliability of the rating is reduced. SUMMARY

[0004] Therefore, it is necessary to provide a financial rating data optimization method, device, computer equipment, computer readable storage medium and computer program product capable of improving the reliability and timeliness of credit quantitative rating.

[0005] In a first aspect, the present application provides a financial rating data optimization method, comprising:

[0006] Collecting financial raw data from associated data sources, performing preprocessing operations on the financial raw data to obtain a standardized clean financial data set; the preprocessing operations include format standardization conversion, missing value filling based on industry average and abnormal value filtering of error data;

[0007] Performing integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification on the clean financial data set in turn, and outputting rule review abnormal items and verification result data;

[0008] Based on the financial statement data and the calculated financial indicators within a preset period, matching the corresponding indicator group according to the preset industry first-level classification label; for each indicator group, calling the corresponding preconfigured indicator group decision matrix, the rows of the indicator group decision matrix representing the specific financial indicators in the indicator group, and the columns representing the threshold intervals of the abnormal degree of indicators; calculating the anti-fraud score of the indicator group through the indicator group decision matrix; and performing weighted average on the anti-fraud scores of multiple indicator groups to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level;

[0009] According to the corresponding risk early warning rule set matched with the manufacturing industry, the wholesale and retail industry and the general industry, the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters;

[0010] Based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed; the bad sample distribution rate of each cell in the comprehensive risk decision matrix is valued to generate a financial screening comprehensive risk level.

[0011] In one of the embodiments, the method further comprises:

[0012] The integrated rule checks the abnormal items, the financial anti-fraud comprehensive score, the financial health degree level and the comprehensive risk level to generate a comprehensive financial data screening report.

[0013] In one of the embodiments, the method further comprises:

[0014] Based on the comprehensive financial data screening report, the rating input data is corrected and the rating model parameters are adjusted;

[0015] The corrected rating input data and the adjusted rating model parameters are pushed to the business terminal.

[0016] In one of the embodiments, the method further comprises:

[0017] Based on the comprehensive financial data screening report, the rating process is suspended or the rating input data is verified when an abnormal data situation is identified; the abnormal data situation includes reaching a preset fraud risk level or reaching a preset financial data inaccuracy level.

[0018] In one of the embodiments, the associated data sources include five types of subject data domains of customer subjects, financial information, customer ratings, credit business and public information, and the financial raw data is structured financial data set output by the processed structured data source.

[0019] In one of the embodiments, the comprehensive financial data screening report further includes the corresponding quantitative evaluation results of the consistency and timeliness of the financial data; wherein the consistency evaluation is the quantitative calculation result of the matching degree of subject data between offline reports and online reports, and the timeliness evaluation is the quantitative determination result of the report update time length and the audit report usage state;

[0020] The integrity check is used to quantitatively check the missing rate of the report data field corresponding to the clean financial data set to determine whether there is serious data missing; the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise's report through data hash comparison; the validity check is used to check whether the report subject data exists obvious value falsification characteristics or abnormal situation of key subject value being 0.

[0021] In a second aspect, the present application also provides an optimization device for financial rating data, comprising:

[0022] a data collection module, configured to collect financial raw data from an associated data source, perform a preprocessing operation on the financial raw data, and obtain a standardized clean financial dataset; the preprocessing operation comprises format standardization conversion, missing value filling based on industry average, and outlier filtering of error data;

[0023] a financial statement strategy rule module, configured to sequentially perform integrity verification, repeatability verification, validity verification, reconciliation relationship verification, and business rule verification on the clean financial dataset, and output rule review abnormal items and verification result data;

[0024] a financial anti-fraud module, configured to match corresponding index groups according to a preset industry first-level classification label based on financial statement data and calculated financial indicators within a preset period; for each index group, a corresponding pre-configured index group decision matrix is called, the rows of the index group decision matrix represent specific financial indicators in the index group, and the columns represent threshold intervals of index abnormality degrees; the index group decision matrix is used to calculate an anti-fraud score of the index group; the anti-fraud scores of the plurality of index groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level;

[0025] a financial health degree module, configured to match corresponding risk warning rule groups according to manufacturing, wholesale and retail industries, and general industry, calculate a financial health degree score and a financial health degree level based on the calculated financial indicators and financial health degree model parameters;

[0026] a financial screening comprehensive score module, configured to construct a comprehensive risk decision matrix based on the financial anti-fraud risk level and the financial health degree level; each cell in the comprehensive risk decision matrix is valued according to a bad sample distribution rate, and a financial screening comprehensive risk level is generated.

[0027] In a third aspect, the present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0028] collecting financial raw data from an associated data source, performing a preprocessing operation on the financial raw data, and obtaining a standardized clean financial dataset; the preprocessing operation comprises format standardization conversion, missing value filling based on industry average, and outlier filtering of error data;

[0029] sequentially performing integrity verification, repeatability verification, validity verification, reconciliation relationship verification, and business rule verification on the clean financial dataset, and outputting rule review abnormal items and verification result data;

[0030] based on the financial statement data and the calculated financial indicators within the preset period, corresponding indicator groups are matched according to preset industry first-level classification labels; for each indicator group, a corresponding pre-configured indicator group decision matrix is called, the rows of the indicator group decision matrix represent specific financial indicators in the indicator group, and the columns represent threshold intervals of the abnormal degree of indicators; through the indicator group decision matrix, the anti-fraud score of the indicator group is calculated; the anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level;

[0031] based on the calculated financial indicators and the financial health degree model parameters, the financial health degree score and the financial health degree level are calculated according to the corresponding risk warning rule groups matched according to the manufacturing industry, the wholesale and retail industry, and the general industry;

[0032] based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed; according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated by assigning values.

[0033] In a fourth aspect, the present application also provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0034] financial raw data is collected from an associated data source, and a preprocessing operation is performed on the financial raw data to obtain a standardized clean financial data set; the preprocessing operation includes format standardization conversion, missing value filling based on the mean value of the same industry, and abnormal value filtering of error data;

[0035] The clean financial data set is sequentially subjected to integrity verification, repeatability verification, validity verification, reconciliation relationship verification, and business rule verification, and the rule review abnormal items and the verification result data are output;

[0036] based on the financial statement data and the calculated financial indicators within the preset period, corresponding indicator groups are matched according to preset industry first-level classification labels; for each indicator group, a corresponding pre-configured indicator group decision matrix is called, the rows of the indicator group decision matrix represent specific financial indicators in the indicator group, and the columns represent threshold intervals of the abnormal degree of indicators; through the indicator group decision matrix, the anti-fraud score of the indicator group is calculated; the anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level;

[0037] based on the calculated financial indicators and the financial health degree model parameters, the financial health degree score and the financial health degree level are calculated according to the corresponding risk warning rule groups matched according to the manufacturing industry, the wholesale and retail industry, and the general industry;

[0038] Based on the financial anti-fraud risk level and the financial health level, a comprehensive risk decision matrix is constructed; according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated.

[0039] In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps:

[0040] Financial raw data is collected from the associated data source, and a preprocessing operation is performed on the financial raw data to obtain a standardized clean financial data set; the preprocessing operation includes format standardization conversion, missing value filling based on the same industry mean, and error data abnormal value filtering;

[0041] The clean financial data set is sequentially subjected to integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification, and the rule review abnormal items and the verification result data are output;

[0042] Based on the financial report data and the calculated financial indicators within a preset period, the corresponding indicator group is matched according to the preset industry first-level classification label; for each indicator group, the corresponding preconfigured indicator group decision matrix is called, the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the abnormal degree of the indicators; through the indicator group decision matrix, the anti-fraud score of the indicator group is calculated; the anti-fraud scores of multiple indicator groups are weighted and averaged to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level;

[0043] According to the corresponding risk warning rule group matched by the manufacturing industry, the wholesale and retail industry and the general industry, the financial health score and the financial health level are calculated based on the calculated financial indicators and the financial health model parameters;

[0044] Based on the financial anti-fraud risk level and the financial health level, a comprehensive risk decision matrix is constructed; according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated.

[0045] The optimization method, device, computer equipment, computer readable storage medium and computer program product of the financial rating data collect financial raw data from an associated data source, perform a preprocessing operation on the financial raw data to obtain a standardized clean financial data set, the preprocessing operation includes format standardization conversion, missing value filling based on industry average and abnormal value filtering of error data, sequentially perform integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification on the clean financial data set, output rule review abnormal items and verification result data, match corresponding index groups according to a preset industry first-level classification label based on financial statement data and calculated financial indicators in a preset period, for each index group, call the corresponding preconfigured index group decision matrix, the rows of the index group decision matrix represent specific financial indicators in the index group, and the columns represent threshold intervals of index abnormality degrees, calculate the anti-fraud score of the index group through the index group decision matrix, perform weighted average on the anti-fraud scores of the plurality of index groups to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level, match corresponding risk warning rule groups according to manufacturing, wholesale and retail industries and general industry, calculate the financial health score and the financial health level based on the calculated financial indicators and the financial health model parameters, construct a comprehensive risk decision matrix based on the financial anti-fraud risk level and the financial health level, and generate a financial screening comprehensive risk level according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix. By performing the preprocessing operation on the collected financial raw data to obtain the standardized clean financial data set, the uniformity and integrity of the data structure are ensured, and the data quality is improved; by performing rule verification on the clean financial data set, the problems of data consistency and timeliness are effectively solved; by performing weighted average on the anti-fraud scores of each index group to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level, false data can be identified, thereby ensuring the reliability of the data; by calculating the financial health score and the financial health level, the financial basis reliability can be ensured; by ensuring the quality of the financial data input into the rating system from multiple dimensions, financial data anomalies and fraud signs can be found in time, high-risk data can be warned in advance, and rating result errors caused by financial data problems are reduced, so that the credit rating is more in line with the actual risk of customers. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application or the related art. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0047] Figure 1An application environment diagram of the financial rating data optimization method in an embodiment;

[0048] Figure 2 A flowchart of the financial rating data optimization method in an embodiment;

[0049] Figure 3 A flowchart of the financial rating data optimization method in another embodiment;

[0050] Figure 4 A structural block diagram of the financial rating data optimization device in an embodiment;

[0051] Figure 5 An internal structure diagram of the computer device in an embodiment. DETAILED DESCRIPTION

[0052] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0053] It should be noted that the terms "first", "second", etc. used in the present application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "include" and "have" used in the present application and any variations thereof are intended to cover non-exclusive inclusion. The term "a plurality of" used in the present application means two or more. The term "and / or" used in the present application means one of the options or any combination of multiple options.

[0054] The financial rating data optimization method provided by the embodiments of the present application can be applied in an application environment as shown in the figure. Figure 1 The terminal 102 communicates with the server 104 through a network. The data storage system can store data required to be processed by the server 104. The data storage system can be integrated on the server 104, or placed on a cloud or other network server. Specifically, the terminal 102 or the server 104 completes a financial rating data optimization method, which includes the following steps.

[0055] The financial raw data is collected from the associated data source, and a preprocessing operation is performed on the financial raw data to obtain a standardized clean financial data set. The preprocessing operation includes format standardization conversion, missing value filling based on industry average, and error data anomaly value filtering.

[0056] The clean financial data set is sequentially subjected to integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification, and the rule review abnormal items and the verification result data are output.

[0057] Based on the financial statement data and the calculated financial indicators within the preset period, the corresponding indicator groups are matched according to the preset industry first-level classification labels; for each indicator group, the corresponding pre-configured indicator group decision matrix is called, the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the abnormal degree of indicators; the anti-fraud score of the indicator group is calculated through the indicator group decision matrix; the anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level;

[0058] According to the manufacturing industry, the wholesale and retail industry, and the general industry, the corresponding risk warning rule group is matched, the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters;

[0059] Based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed; according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated by assigning values.

[0060] The terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, unmanned aerial vehicles, low-altitude aerial vehicles, Internet of Things devices, and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle device, a projection device, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0061] In an exemplary embodiment, as shown in Figure 2 , a financial rating data optimization method is provided. The method is applied to the server in Figure 1 for illustration, including the following steps 202 to 210. Among them:

[0062] Step 202, collect financial raw data from associated data sources, perform preprocessing operations on the financial raw data, and obtain a standardized clean financial data set; the preprocessing operation includes format standardization conversion, missing value filling based on the average of the same industry, and outlier filtering of error data.

[0063] The format standardization conversion refers to standardizing or formatting the financial raw data, so as to eliminate the structural inconsistency caused by different data sources and different input standards, and ensure the uniformity of the data in format, unit and coding. The missing value filling based on the average of the same industry refers to processing the blank or missing records in the data by using the context-aware interpolation method. The abnormal value filtering of error data refers to identifying and processing the data points that are significantly different from the behavior of most data.

[0064] Exemplarily, the associated data source includes five types of subject data domains of customer subject, financial information, customer rating, credit business and public information, and the financial raw data is a structured financial data set output by a structured data source after processing and backflow.

[0065] Exemplarily, five types of subject data domains including customer subject, financial information, customer rating, credit business and public information are collected from the associated data source, and the collected data is processed and backflowed into a structured data source. The structured financial data set output by the structured data source is used to obtain the financial raw data, and the pre-processing operations including format standardization conversion, missing value filling based on the average of the same industry and abnormal value filtering of error data are performed on the financial raw data to obtain the standardized clean financial data set.

[0066] In step 204, the clean financial data set is sequentially subjected to integrity check, repeatability check, validity check, reconciliation relationship check and business rule check, and the rule review abnormal items and the check result data are output.

[0067] The integrity check refers to checking whether there is a serious data missing in the data set. The repeatability check refers to checking whether there is a case that the enterprise uses the past financial data of the enterprise or uses the financial data of other enterprises for submission in the data set. The validity check refers to checking whether there is a significant fraud behavior or a case that the key subject is 0 in the subject data in the data set. The reconciliation relationship check refers to checking whether there is a logical consistency and correlation between different financial data. The business rule check refers to calculating the business check rules based on the calculated financial indicators, to evaluate the health degree, compliance and whether there is a major misstatement risk of the enterprise financial status, and storing the rules triggered by the business check and the related prompt information.

[0068] Exemplarily, based on the obtained standardized clean financial data set, the data set is sequentially subjected to integrity check, repeatability check, validity check, reconciliation relationship check and business rule check, to detect whether there are report data missing, report duplication, reconciliation failure, unreasonable cost and expense structure, uncoordinated asset and liability subjects, and mismatched equity and profit, and output the rule review abnormal items and the check result data.

[0069] At step 206, based on the financial statement data and the calculated financial indicators within the preset period, the corresponding indicator groups are matched according to the preset industry first-level classification label; for each indicator group, the corresponding pre-configured indicator group decision matrix is called, the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the abnormal degree of the indicators; the anti-fraud score of the indicator group is calculated through the indicator group decision matrix; the anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level.

[0070] The financial indicators include profitability indicators, solvency indicators, operating capacity indicators, cash flow indicators, and market value indicators. The preset period can be one year, which is not limited herein. The preset industry first-level classification label can be a national standard first-level industry.

[0071] Exemplarily, at least one year of financial statement data of an enterprise is obtained, and financial indicators are calculated according to the financial statement data. Based on the obtained financial statement data and the calculated financial indicators, corresponding multiple indicator groups are matched according to the preset industry first-level classification label, so that the indicators and benchmarks that need to be concerned in the industry to which the enterprise belongs can be found. For each indicator group, the corresponding pre-configured indicator group decision matrix is called, wherein the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the abnormal degree of the indicators, and the indicator groups and the combinations of the indicator threshold values used in different industries are different. The anti-fraud scores of each indicator group are calculated through the indicator group decision matrix, and the anti-fraud scores of the various indicator groups are weighted and averaged, so that the financial anti-fraud comprehensive score and the financial anti-fraud risk level are obtained.

[0072] At step 208, the corresponding risk early warning rule groups are matched according to the manufacturing industry, the wholesale and retail industry, and the general industry, and the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters.

[0073] The general industry refers to the remaining industries other than the manufacturing industry, the wholesale and retail industry, the financial industry, education, public management, social security and social organizations, and international organizations.

[0074] The financial health degree model parameters include various parameters for measuring the profitability of the enterprise, various parameters for measuring the solvency of the enterprise, various parameters for measuring the operating capacity of the enterprise, various parameters for measuring the cash flow of the enterprise, and various parameters for measuring the development potential of the enterprise.

[0075] Exemplarily, the corresponding risk early warning rule groups are matched according to the manufacturing industry, the wholesale and retail industry, and the general industry, and the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters according to the calculation logic of the corresponding risk early warning rules.

[0076] In step 210, a comprehensive risk decision matrix is constructed based on the financial anti-fraud risk level and the financial health level; and a financial screening comprehensive risk level is generated by assigning values to each cell in the comprehensive risk decision matrix according to the bad sample distribution rate of each cell.

[0077] By way of example, the comprehensive risk decision matrix is constructed, the obtained financial health level is taken as the horizontal axis of the comprehensive risk decision matrix, the obtained financial anti-fraud risk level is taken as the vertical axis of the comprehensive risk decision matrix, and each cell in the comprehensive risk decision matrix is assigned a financial screening comprehensive risk level according to the bad sample distribution rate of each cell, so as to obtain a financial screening comprehensive score.

[0078] In the above financial rating data optimization method, the collected financial raw data is pre-processed to obtain a standardized clean financial data set, thereby ensuring the uniformity and integrity of the data structure and improving the data quality; the clean financial data set is subjected to rule checking to effectively solve the problems of data consistency and timeliness; the anti-fraud scores of each index group are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level, which can identify false data and thus ensure the reliability of the data; the financial health degree score and the financial health degree level are calculated to ensure the reliability of the financial basis; the quality of the financial data input into the rating system is ensured from multiple dimensions, and abnormal financial data and fraud signs are discovered in a timely manner, early warning of high-risk data is provided, and rating result errors caused by financial data problems are reduced, so that the credit rating is more in line with the actual risk of the customer.

[0079] In one embodiment, the method further comprises: integrating the rule checking abnormal items, the financial anti-fraud comprehensive score, the financial health level, and the comprehensive risk level to generate a comprehensive financial data screening report.

[0080] By way of example, the rule checking abnormal items, the financial anti-fraud comprehensive score, the financial health level, and the financial screening comprehensive risk level obtained after checking the clean financial data set are used to generate a comprehensive financial data screening report, which includes the evaluation of the quality of the financial data, i.e., whether the consistency and timeliness meet the standards, the fraud risk level, the financial health level, and the like.

[0081] In this embodiment, the comprehensive financial data screening report is generated by integrating the checking and rating results, which can provide a multi-dimensional financial status, which is not only beneficial for credit rating, but also applicable to credit approval, post-loan management, customer stratification maintenance, and other business scenarios.

[0082] In one embodiment, the method further comprises: based on the comprehensive financial data screening report, revising the rating input data and adjusting the rating model parameters; and pushing the revised rating input data and the adjusted rating model parameters to a business terminal.

[0083] Exemplarily, the comprehensive financial data screening report is fed back to the intelligent rating system, so that the rating input data is corrected and the rating model parameters are adjusted according to the comprehensive financial data screening report, and the corrected rating input data and the adjusted rating model parameters are pushed to the rating personnel or the risk control personnel at the business end.

[0084] In this embodiment, the intelligent rating system performs data correction and parameter adjustment based on the comprehensive financial data screening report and pushes the data to the business end, which can standardize rating data management and improve business efficiency, thereby promoting the standardization of credit rating business processes.

[0085] In one embodiment, the method further comprises: suspending the execution of the rating process or verifying the rating input data based on the comprehensive financial data screening report identifying an abnormal data situation; the abnormal data situation includes reaching a preset fraud risk level or reaching a preset financial data inaccuracy level.

[0086] Exemplarily, in the case where the comprehensive financial data screening report identifies an abnormal data situation that the data reaches a preset fraud risk level or reaches a preset financial data inaccuracy level, the execution of the rating process is suspended or the rating input data is verified, and the like.

[0087] In this embodiment, the abnormal data identified based on the comprehensive financial data screening report triggers the rating suspension, data verification and other risk control processes, which can improve the reliability and timeliness of credit quantitative rating.

[0088] In one embodiment, the comprehensive financial data screening report further includes quantitative evaluation results of consistency and timeliness of the financial data; wherein the consistency evaluation is a quantitative calculation result of the matching degree of subject data between offline reports and online reports, and the timeliness evaluation is a quantitative determination result of the report update duration and the audit report usage state; the integrity check is used to quantitatively check the missing rate of report data fields corresponding to the clean financial data set to determine whether there is serious data missing; the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise's report through data hash comparison; the validity check is used to check whether the report subject data exists obvious value falsification characteristics or abnormal situation of key subject value being 0.

[0089] The data hash comparison includes integrity check of the same data at different time points, or consistency check of different data.

[0090] Exemplarily, the consistency of the financial data is evaluated to obtain a matching quantitative calculation result of subject data between offline reports and online reports, so as to reduce the deviation of the rating result calculated based on the error data from the true risk due to the failure of the subject data to correspond; the timeliness of the financial data is evaluated to obtain a quantitative determination result of the report update time length and the audit report use state, so as to effectively solve the problem that the rating result cannot reflect the actual risk change of the customer due to the failure of the report data to be updated.

[0091] Exemplarily, the integrity of the financial data is verified to quantify the missing rate of the report data field corresponding to the clean financial data set, so as to determine whether there is a serious data missing; the repeatability of the financial data is verified by data hash comparison to verify whether the report exists the situation of reusing the historical report of the enterprise itself or the report of other enterprises; the validity of the financial data is verified to verify whether the subject data of the report exists obvious value fraud characteristics or abnormal situation that the key subject value is 0.

[0092] In this embodiment, by obtaining the quantitative evaluation results of the consistency and timeliness of the financial data, potential financial risk points can be efficiently identified, data quality can be guaranteed, and the authenticity and accuracy of the credit quantitative rating can be improved.

[0093] As shown in Figure 3 , a specific embodiment is used to illustrate the optimization method of the financial rating data, including steps 302 to 314. Among them,

[0094] Step 302, collecting financial raw data from associated data sources, performing preprocessing operation on the financial raw data to obtain standardized clean financial data set.

[0095] Specifically, five types of theme data domains including customer subject, financial information, customer rating, credit business and public information are collected from associated data sources, and the collected data is processed to return to a structured data source. The structured financial data set output by the structured data source is used to obtain the financial raw data, and the financial raw data is preprocessed including format standardization conversion, missing value filling based on industry average and abnormal value filtering of error data, to obtain the standardized clean financial data set.

[0096] Step 304, sequentially performing integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification on the standardized clean financial data set.

[0097] Specifically, based on the obtained standardized clean financial data set, integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification are sequentially performed on the data set to detect whether there are problems such as report data missing, report duplication, reconciliation failure, unreasonable cost and expense structure, uncoordinated asset and liability accounts, and mismatched equity and profit in the data set, and output rule checking abnormal items and verification result data.

[0098] Step 306, based on the financial statement data and the calculated financial indicators within a preset period, obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level.

[0099] Specifically, at least one year of financial statement data of an enterprise is obtained, and financial indicators are calculated according to the financial statement data. Based on the obtained financial statement data and the calculated financial indicators, a plurality of index groups corresponding to a preset industry first-level classification label are matched, so that the indexes and benchmarks that need to be concerned in the industry to which the enterprise belongs can be found. For each index group, the corresponding pre-configured index group decision matrix is called, wherein the rows of the index group decision matrix represent the specific financial indicators in the index group, and the columns represent the threshold intervals of the index abnormality degree, and the index groups and index threshold combinations used by different industries are different. Through the index group decision matrix, the anti-fraud score of each index group is calculated, and the anti-fraud scores of the index groups are weighted and averaged, so as to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level.

[0100] Among them, the index threshold needs to be parameterized, and for different index groups used by different industries, the index thresholds corresponding to the index groups are stored in the database, so as to facilitate the modification and maintenance of the index thresholds.

[0101] Step 308, based on the calculated financial indicators and the financial health degree model parameters, calculate the financial health degree score and the financial health degree level.

[0102] Specifically, the corresponding risk early warning rule groups are matched according to the manufacturing industry, the wholesale and retail industry and the general industry, and based on the calculated financial indicators and the financial health degree model parameters, the calculation logic of the corresponding risk early warning rules is calculated, so as to obtain the financial health degree score and the financial health degree level.

[0103] Step 310, based on the financial anti-fraud risk level and the financial health degree level, generate a financial screening comprehensive risk level.

[0104] Specifically, the comprehensive risk decision matrix is constructed, the financial health degree grade obtained is taken as the horizontal axis of the comprehensive risk decision matrix, the financial anti-fraud risk grade obtained is taken as the vertical axis of the comprehensive risk decision matrix, each cell in the comprehensive risk decision matrix is given a financial screening comprehensive risk grade according to the bad sample distribution rate of the cell, and thus a financial screening comprehensive score is obtained.

[0105] In step 312, the rule checking abnormal item, the financial anti-fraud comprehensive score, the financial health degree grade and the comprehensive risk grade are integrated to generate a comprehensive financial data screening report.

[0106] Specifically, according to the rule checking abnormal item, the financial anti-fraud comprehensive score, the financial health degree grade and the financial screening comprehensive risk grade obtained after the clean financial data set is verified, a comprehensive financial data screening report is generated, which includes the financial data quality evaluation, i.e., whether the data consistency, timeliness, fraud risk grade and financial health degree meet the standards.

[0107] In step 314, the comprehensive financial data screening report is fed back to the intelligent rating system.

[0108] Specifically, the comprehensive financial data screening report is fed back to the intelligent rating system, so that the rating input data and the rating model parameters are corrected according to the comprehensive financial data screening report, and the corrected rating input data and the adjusted rating model parameters are pushed to the rating personnel or the risk control personnel at the business end.

[0109] In the case where the comprehensive financial data screening report identifies that the data reaches a preset fraud risk level or reaches a preset financial data inaccuracy level, the rating process is suspended or the rating input data is verified.

[0110] It should be understood that although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps. It can be understood that the steps in different embodiments can be freely combined as needed, and various non-contradictory schemes formed by the combination are within the scope of protection of the present application.

[0111] Based on the same inventive concept, the embodiments of the present application also provide a financial rating data optimization device for implementing the financial rating data optimization method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more financial rating data optimization device embodiments provided below can refer to the limitations of the financial rating data optimization method in the above, which will not be described here.

[0112] In one exemplary embodiment, as shown in Figure 4 A financial rating data optimization device 400 is provided, comprising a data acquisition module 402, a financial statement strategy rule module 404, a financial fraud prevention module 406, a financial health degree module 408, and a financial screening comprehensive score module 410, wherein:

[0113] The data acquisition module 402 is configured to acquire financial raw data from an associated data source, and perform a preprocessing operation on the financial raw data to obtain a standardized clean financial data set. The preprocessing operation includes format standardization conversion, missing value filling based on industry average, and outlier filtering of error data.

[0114] The financial statement strategy rule module 404 is configured to sequentially perform integrity verification, repeatability verification, validity verification, reconciliation relationship verification, and business rule verification on the clean financial data set, and output rule review abnormal items and verification result data.

[0115] The financial fraud prevention module 406 is configured to match the corresponding index group according to the preset industry first-level classification label based on the financial statement data and the calculated financial indicators within a preset period. For each index group, the corresponding pre-configured index group decision matrix is called. The rows of the index group decision matrix represent specific financial indicators in the index group, and the columns represent threshold intervals of index abnormality degree. Through the index group decision matrix, the anti-fraud score of the index group is calculated. The anti-fraud scores of multiple index groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level.

[0116] The financial health degree module 408 is configured to match the corresponding risk warning rule group according to the manufacturing industry, the wholesale and retail industry, and the general industry, and calculate the financial health degree score and the financial health degree level based on the calculated financial indicators and the financial health degree model parameters.

[0117] The financial screening comprehensive score module 410 is configured to construct a comprehensive risk decision matrix based on the financial fraud risk level and the financial health degree level. According to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated by assigning values.

[0118] In one of the embodiments, the financial screening comprehensive scoring module is further configured to integrate the rule checking abnormal items, the financial anti-fraud comprehensive score, the financial health level and the comprehensive risk level to generate a comprehensive financial data screening report.

[0119] In one of the embodiments, the financial screening comprehensive scoring module is further configured to correct the rating input data and adjust the rating model parameters based on the comprehensive financial data screening report, and push the corrected rating input data and the adjusted rating model parameters to the business terminal.

[0120] In one of the embodiments, the financial screening comprehensive scoring module is further configured to suspend the rating process or verify the rating input data when an abnormal data situation is identified based on the comprehensive financial data screening report, wherein the abnormal data situation includes reaching a preset fraud risk level or reaching a preset financial data inaccuracy level.

[0121] In one of the embodiments, the data collection module is further configured to associate the data sources with five types of subject data domains including customer subjects, financial information, customer ratings, credit business and public information, and the financial raw data is a structured financial data set output by a processed backflow structured data source.

[0122] In one of the embodiments, the financial screening comprehensive scoring module is further configured to include the corresponding quantitative evaluation results of the consistency and timeliness of the financial data in the comprehensive financial data screening report, wherein the consistency evaluation is a quantitative calculation result of the matching degree of subject data between offline reports and online reports, the timeliness evaluation is a quantitative determination result of the report update time length and the audit report usage state, the integrity check is used to quantitatively check the missing rate of the report data field corresponding to the clean financial data set to determine whether there is a serious data missing, the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise's report through data hash comparison, and the validity check is used to check whether the report subject data exists the abnormal situation of obvious value falsification characteristics or key subject value being 0.

[0123] The above-mentioned modules in the financial rating data optimization device can be realized by software, hardware and their combinations in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above-mentioned modules by the processor.

[0124] In one of the embodiments, a computer device is provided, which can be a terminal, and the internal structure diagram thereof can be as shown in Figure 5The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to perform wired or wireless communication with external terminals. The wireless communication can be realized through WIFI, mobile cellular network, near field communication (NFC) or other technologies. The computer program is executed by the processor to implement a financial rating data optimization method.

[0125] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0126] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0127] Raw financial data is collected from an associated data source, and a preprocessing operation is performed on the raw financial data to obtain a standardized clean financial data set. The preprocessing operation includes format standardization conversion, missing value filling based on industry average, and outlier filtering of error data.

[0128] The clean financial data set is sequentially subjected to integrity verification, repeatability verification, validity verification, reconciliation relationship verification and business rule verification, and rule review abnormal items and verification result data are output.

[0129] Based on the financial statement data and the calculated financial indicators within a preset period, the corresponding indicator groups are matched according to the preset industry first-level classification labels. For each indicator group, the corresponding preconfigured indicator group decision matrix is retrieved, the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the indicator abnormality degree. Through the indicator group decision matrix, the anti-fraud score of the indicator group is calculated. The anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level.

[0130] According to the corresponding risk early warning rule set matched with the manufacturing industry, the wholesale and retail industry and the general industry, the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters;

[0131] Based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed; and according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, a financial screening comprehensive risk level is generated.

[0132] In one embodiment, when the processor executes the computer program, the following steps are also implemented: integrating the rule check abnormal item, the financial anti-fraud comprehensive score, the financial health degree level and the comprehensive risk level to generate a comprehensive financial data screening report.

[0133] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on the comprehensive financial data screening report, the rating input data is corrected and the rating model parameters are adjusted; and the corrected rating input data and the adjusted rating model parameters are pushed to the business terminal.

[0134] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on the comprehensive financial data screening report, the rating input data is corrected and the rating model parameters are adjusted; and the corrected rating input data and the adjusted rating model parameters are pushed to the business terminal.

[0135] In one embodiment, when the processor executes the computer program, the following steps are also implemented: the associated data sources include five types of subject data domains of customer subjects, financial information, customer ratings, credit business and public information, and the financial raw data is a structured financial data set output by a processed backflow structured data source.

[0136] In one embodiment, when the processor executes the computer program, the following steps are also implemented: the comprehensive financial data screening report also includes the corresponding quantitative evaluation results of the consistency and timeliness of the financial data; wherein the consistency evaluation is the quantitative calculation result of the matching degree of subject data between offline reports and online reports, and the timeliness evaluation is the quantitative determination result of the report update time length and the audit report usage state; the integrity check is used to quantitatively check the missing rate of the report data field corresponding to the clean financial data set to determine whether there is a serious data missing; the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise's report through data hash comparison; and the validity check is used to check whether the report subject data exists obvious value falsification characteristics or abnormal situation of key subject value being 0.

[0137] The implementation principles and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be described here again.

[0138] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the following steps:

[0139] Raw financial data is collected from a related data source, and a preprocessing operation is performed on the raw financial data to obtain a standardized clean financial data set. The preprocessing operation includes format standardization conversion, missing value filling based on industry average, and outlier filtering of error data.

[0140] The clean financial data set is sequentially subjected to integrity verification, repeatability verification, validity verification, reconciliation relationship verification, and business rule verification, and rule review abnormal items and verification result data are output.

[0141] Based on the financial statement data and the calculated financial indicators within a preset period, the corresponding indicator group is matched according to the preset industry first-level classification label. For each indicator group, the corresponding preconfigured indicator group decision matrix is called. The rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the indicator abnormality degree. Through the indicator group decision matrix, the anti-fraud score of the indicator group is calculated. The anti-fraud scores of multiple indicator groups are weighted and averaged to obtain a financial anti-fraud comprehensive score and a financial anti-fraud risk level.

[0142] According to the corresponding risk warning rule group matched according to the manufacturing industry, the wholesale and retail industry, and the general industry, the financial health degree score and the financial health degree level are calculated based on the calculated financial indicators and the financial health degree model parameters.

[0143] Based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed. According to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated.

[0144] In one embodiment, the computer program is executed by the processor to further implement the following steps: integrating the rule review abnormal items, the financial anti-fraud comprehensive score, the financial health degree level, and the comprehensive risk level, and generating a comprehensive financial data screening report.

[0145] In one embodiment, the computer program is executed by the processor to further implement the following steps: based on the comprehensive financial data screening report, correcting the rating input data and adjusting the rating model parameters; and pushing the corrected rating input data and the adjusted rating model parameters to a business terminal.

[0146] In one embodiment, the computer program, when executed by the processor, further implements the following steps: suspending the rating process or verifying the rating input data when an abnormal data situation is identified based on the comprehensive financial data screening report; the abnormal data situation includes reaching a preset fraud risk level or reaching a preset financial data inaccuracy level.

[0147] In one embodiment, the computer program, when executed by the processor, further implements the following steps: the associated data sources include five types of subject data domains of customer subjects, financial information, customer ratings, credit services, and public information, and the financial raw data is a structured financial data set output by a processed backflow structured data source.

[0148] In one embodiment, the computer program, when executed by the processor, further implements the following steps: the comprehensive financial data screening report further includes quantitative evaluation results of consistency and timeliness of financial data; wherein the consistency evaluation is a quantitative calculation result of the matching degree of subject data between offline reports and online reports, and the timeliness evaluation is a quantitative determination result of report update time length and audit report usage state; the integrity check is used to quantitatively check the missing rate of report data fields corresponding to the clean financial data set to determine whether there is a serious data missing; the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise report through data hash comparison; the validity check is used to check whether the report subject data exists obvious value falsification characteristics or abnormal situation of key subject value being 0.

[0149] The implementation principles and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.

[0150] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps:

[0151] Collecting financial raw data from associated data sources, performing preprocessing operations on the financial raw data to obtain a standardized clean financial data set; the preprocessing operations include format standardization conversion, missing value filling based on industry average, and abnormal value filtering of error data;

[0152] Performing integrity check, repeatability check, validity check, reconciliation relationship check and business rule check on the clean financial data set in sequence, and outputting rule review abnormal items and check result data;

[0153] Based on the financial statement data and the calculated financial indicators within the preset period, the corresponding indicator groups are matched according to the preset industry first-level classification labels; for each indicator group, the corresponding pre-configured indicator group decision matrix is called, the rows of the indicator group decision matrix represent the specific financial indicators in the indicator group, and the columns represent the threshold intervals of the abnormal degree of the indicators; the anti-fraud score of the indicator group is calculated through the indicator group decision matrix; the anti-fraud scores of the multiple indicator groups are weighted and averaged to obtain the financial anti-fraud comprehensive score and the financial anti-fraud risk level;

[0154] According to the manufacturing industry, the wholesale and retail industry, and the general industry, the corresponding risk warning rule group is matched, and based on the calculated financial indicators and the financial health degree model parameters, the financial health degree score and the financial health degree level are calculated;

[0155] Based on the financial anti-fraud risk level and the financial health degree level, a comprehensive risk decision matrix is constructed; according to the bad sample distribution rate of each cell in the comprehensive risk decision matrix, the financial screening comprehensive risk level is generated by assigning values.

[0156] In one embodiment, the computer program is executed by the processor to further implement the following steps: integrating the rule checking abnormal items, the financial anti-fraud comprehensive score, the financial health degree level and the comprehensive risk level, and generating a comprehensive financial data screening report.

[0157] In one embodiment, the computer program is executed by the processor to further implement the following steps: based on the comprehensive financial data screening report, correcting the rating input data and adjusting the rating model parameters; the corrected rating input data and the adjusted rating model parameters are pushed to the business terminal.

[0158] In one embodiment, the computer program is executed by the processor to further implement the following steps: based on the comprehensive financial data screening report, correcting the rating input data and adjusting the rating model parameters; the corrected rating input data and the adjusted rating model parameters are pushed to the business terminal.

[0159] In one embodiment, the computer program is executed by the processor to further implement the following steps: the associated data sources include five types of subject data domains of customer subjects, financial information, customer ratings, credit business, and public information, and the financial raw data is a structured financial data set output by a structured data source after processing and returning.

[0160] In one embodiment, the computer program, when executed by the processor, further implements the following steps: the comprehensive financial data screening report further includes quantitative evaluation results corresponding to consistency and timeliness of the financial data; wherein the consistency evaluation is a quantitative calculation result of matching degree between offline reports and online reports, and the timeliness evaluation is a quantitative determination result of report update duration and audit report usage state; the integrity check is used to quantitatively check the missing rate of report data fields corresponding to the clean financial data set, to determine whether there is serious data missing; the repeatability check is used to check whether the report exists the situation of reusing the enterprise's own historical report or other enterprise report through data hash comparison; and the validity check is used to check whether the report subject data exists obvious value falsification characteristics or abnormal situation of key subject value being 0.

[0161] The implementation principles and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be described here again.

[0162] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0163] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0164] The technical features of the above embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0165] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for optimizing financial rating data, characterized in that, The method includes: Raw financial data is collected from related data sources, and preprocessing operations are performed on the raw financial data to obtain a standardized and clean financial dataset. The preprocessing operations include format standardization conversion, missing value imputation based on industry averages, and outlier filtering of erroneous data. The clean financial dataset is sequentially subjected to integrity checks, duplicate checks, validity checks, cross-reference checks, and business rule checks, and the rule check anomalies and check result data are output. Based on financial statement data and calculated financial indicators within a preset time period, corresponding indicator groups are matched according to preset industry primary classification labels; for each indicator group, a corresponding pre-configured indicator group decision matrix is ​​retrieved, where the rows of the indicator group decision matrix represent specific financial indicators within the indicator group, and the columns represent the threshold range of the indicator's abnormality level; the anti-fraud score of the indicator group is calculated through the indicator group decision matrix; and the anti-fraud scores of multiple indicator groups are weighted and averaged to obtain a comprehensive financial anti-fraud score and a financial anti-fraud risk level. The risk warning rule groups are matched according to the manufacturing, wholesale and retail and general industries. Based on the calculated financial indicators and financial health model parameters, the financial health score and financial health level are calculated. Based on the financial anti-fraud risk level and financial health level, a comprehensive risk decision matrix is ​​constructed; values ​​are assigned to each cell in the comprehensive risk decision matrix according to the bad sample distribution rate, and a comprehensive financial identification risk level is generated.

2. The method according to claim 1, characterized in that, The method further includes: The system integrates rule-based checks for anomalies, comprehensive financial fraud prevention scores, financial health levels, and overall risk levels to generate a comprehensive financial data verification report.

3. The method according to claim 2, characterized in that, The method further includes: Based on the comprehensive financial data screening report, the rating input data was corrected and the rating model parameters were adjusted. The revised rating input data and adjusted rating model parameters are pushed to the business terminal.

4. The method according to claim 2, characterized in that, The method further includes: If abnormal data is identified based on the comprehensive financial data verification report, the rating process will be suspended or the rating input data will be verified; the abnormal data situations include reaching the preset fraud risk level or reaching the preset degree of financial data inaccuracy.

5. The method according to claim 1, characterized in that, The associated data source includes five subject data domains: customer entity, financial information, customer rating, credit business, and public information. The original financial data is a structured financial dataset output by a structured data source after processing and backflow.

6. The method according to claim 1, characterized in that, The comprehensive financial data verification report also includes quantitative assessment results of the consistency and timeliness of the financial data; among which, the consistency assessment is the quantitative calculation result of the matching of account data between offline reports and online reports, and the timeliness assessment is the quantitative judgment result of the report update time and the status of the audit report usage. The integrity check is used to quantitatively check the missing field rate of the report data corresponding to the clean financial dataset in order to determine whether there is serious data missing; the duplication check is used to check whether the report reuses the company's own historical reports or reports of other companies through data hash comparison; the validity check is used to check whether the report subject data has obvious numerical fraud characteristics or abnormal situations where the key subject value is 0.

7. A device for optimizing financial rating data, characterized in that, The device includes: The data acquisition module is used to collect raw financial data from related data sources, perform preprocessing operations on the raw financial data, and obtain a standardized and clean financial dataset. The preprocessing operations include format standardization conversion, missing value imputation based on the industry average, and outlier filtering of erroneous data. The financial statement strategy rules module is used to sequentially perform integrity checks, duplicate checks, validity checks, cross-reference checks, and business rule checks on the clean financial dataset, and output rule check anomalies and check result data; The financial anti-fraud module is used to match corresponding indicator groups based on financial statement data and calculated financial indicators within a preset time period, according to preset industry primary classification labels; for each indicator group, it retrieves the corresponding pre-configured indicator group decision matrix, where the rows of the indicator group decision matrix represent specific financial indicators within the indicator group, and the columns represent the threshold range of the indicator's abnormality level; it calculates the anti-fraud score of the indicator group through the indicator group decision matrix; and it performs a weighted average of the anti-fraud scores of multiple indicator groups to obtain a comprehensive financial anti-fraud score and a financial anti-fraud risk level. The Financial Health module is used to match corresponding risk warning rule groups by manufacturing, wholesale and retail and general industries. Based on the calculated financial indicators and financial health model parameters, it calculates the financial health score and financial health level. The financial screening comprehensive scoring module is used to construct a comprehensive risk decision matrix based on the financial anti-fraud risk level and financial health level; and to assign values ​​to each cell in the comprehensive risk decision matrix according to the bad sample distribution rate to generate the comprehensive financial screening risk level.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images