Enterprise credit risk determination method and device, electronic equipment, medium and product

Abstract

The embodiment of the invention provides an enterprise credit risk determination method and device, electronic equipment, a medium and a product. The method comprises the following steps: obtaining a plurality of evaluation indexes corresponding to a to-be-evaluated enterprise and characteristic values corresponding to the evaluation indexes, wherein the plurality of evaluation indexes comprise an evaluation index of an enterprise dimension, an evaluation index of an industry dimension and an evaluation index of a region dimension; based on a combinatorial optimization algorithm, screening out a plurality of target evaluation indexes from the plurality of evaluation indexes; according to the multiple target evaluation indexes and the characteristic values corresponding to the target evaluation indexes, the enterprise credit risk corresponding to the to-be-evaluated enterprise is determined, the limitation of traditional financial index evaluation is broken through, forward-looking indexes are introduced, various factors influencing the enterprise credit risk can be captured more comprehensively, and the enterprise credit risk evaluation efficiency is improved. And indexes with low credit risk distinction degree or large noise are effectively eliminated, repeated information calculation is avoided, the risk identification accuracy is further improved, and the misjudgment risk is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a method and device for determining enterprise credit risk, electronic equipment, medium and product. BACKGROUND

[0002] Enterprise credit risk assessment is the core link of credit decision-making of banks and financial institutions, and the accuracy and scientificity of its results are directly related to the quality and safety of credit assets. An efficient and reliable risk assessment system can effectively identify the credit status of enterprises and accurately distinguish between high-quality customers and high-risk customers, thereby optimizing credit resource allocation while controlling bad debt risk, which is of great significance to the stability of the financial system and the promotion of capital services to the real economy.

[0003] At present, the traditional enterprise credit risk assessment in the industry mainly relies on quantitative analysis based on financial indicators. First, a series of financial ratios reflecting the solvency, profitability, operational capacity and development capacity of enterprises are selected as evaluation indicators, such as asset-liability ratio, liquidity ratio, total asset turnover rate, etc.; then, the experts' experience or simple statistical methods are used to assign certain weights to each indicator; finally, a comprehensive credit score is calculated by weighted summation or weighted average, but the comprehensive credit score determined by the above method has the problem of low accuracy. SUMMARY

[0004] The embodiments of the present application provide a method and device for determining enterprise credit risk, electronic equipment, medium and product, to improve the accuracy of enterprise credit risk assessment.

[0005] In a first aspect, the embodiments of the present application provide a method for determining enterprise credit risk, which comprises:

[0006] obtaining a plurality of evaluation indicators corresponding to the enterprise to be evaluated and feature values corresponding to each evaluation indicator, wherein the plurality of evaluation indicators include evaluation indicators of enterprise dimension, industry dimension and region dimension;

[0007] selecting a plurality of target evaluation indicators from the plurality of evaluation indicators based on a combinatorial optimization algorithm;

[0008] determining the enterprise credit risk of the enterprise to be evaluated according to the plurality of target evaluation indicators and the feature values corresponding to each target evaluation indicator.

[0009] Optionally, the evaluation indexes of the region dimension include a regional economic development index, a regional policy support index, and a regional credit ecological index, the regional economic development index is used to indicate the overall management situation of the region where the enterprise to be evaluated is located, the regional policy support index is used to indicate the credit, industrial policy orientation and strength of the region where the enterprise to be evaluated is located, and the regional credit ecological index is used to indicate the overall credit ecological situation of the region where the enterprise to be evaluated is located.

[0010] Optionally, based on the combinatorial optimization algorithm, a plurality of target evaluation indexes are selected from a plurality of evaluation indexes, including:

[0011] Define decision variables , wherein represents selecting the i-th evaluation index, represents not selecting the i-th evaluation index;

[0012] Define the objective function , wherein is used to indicate the correlation between the evaluation index i and the evaluation index j; wherein n represents the total number of the plurality of evaluation indexes;

[0013] Define the constraint condition , limit the selection of at most m evaluation indexes;

[0014] Define the total QUBO model , wherein is a preset penalty coefficient;

[0015] Determine the evaluation index combination corresponding to the minimum value of H based on the QAOA algorithm;

[0016] According to the evaluation index combination, determine a plurality of target evaluation indexes.

[0017] Optionally, the number of the enterprises to be evaluated is a plurality, and according to the plurality of target evaluation indexes and the characteristic values corresponding to each target evaluation index, the enterprise credit risk corresponding to the enterprise to be evaluated is determined, including:

[0018] According to the characteristic values corresponding to each target evaluation index corresponding to a plurality of enterprises to be evaluated respectively, a plurality of principal components are extracted from the plurality of target evaluation indexes, and scores corresponding to each principal component corresponding to a plurality of enterprises to be evaluated respectively are determined, wherein the proportion of the explained variance of the plurality of principal components is greater than a preset threshold;

[0019] For any enterprise to be evaluated, according to the scores corresponding to each principal component corresponding to the enterprise to be evaluated, the credit comprehensive score corresponding to the enterprise to be evaluated is calculated by using the following formula:

[0020]

[0021] wherein, represents the credit comprehensive score corresponding to the pth enterprise to be evaluated, represents the weight corresponding to the qth principal component, k represents the total number of principal components, represents the corresponding score of the qth principal component of the pth enterprise to be evaluated;

[0022] wherein, the weight wherein represents the eigenvalue corresponding to the qth principal component;

[0023] According to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated, the credit index corresponding to each enterprise to be evaluated is calculated.

[0024] According to the credit index corresponding to each enterprise to be evaluated, the enterprise credit risk corresponding to each enterprise to be evaluated is determined.

[0025] Optionally, according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated, the credit index corresponding to each enterprise to be evaluated is calculated, including:

[0026] Based on the K-means clustering algorithm, according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated, the credit risk level corresponding to each enterprise to be evaluated is determined;

[0027] For any enterprise to be evaluated, according to the credit risk level corresponding to the enterprise to be evaluated, the credit availability index corresponding to the enterprise to be evaluated is determined through the following formula; wherein, represents the credit availability index corresponding to the pth enterprise to be evaluated, represents the credit risk level corresponding to the enterprise to be evaluated; wherein r is the number of credit risk levels; represents the weight corresponding to the credit risk level l;

[0028]

[0029] According to the credit availability index corresponding to the enterprise to be evaluated, the credit investment scale index corresponding to the enterprise to be evaluated is determined through the following formula; wherein, is used to indicate the size of the pth enterprise to be evaluated;

[0030]

[0031] The credit investment scale index corresponding to the enterprise to be evaluated is determined through the formula , wherein, is the estimated value of the default loss rate.

[0032] Optionally, is determined according to a credit business orientation, wherein the credit business orientation comprises: an aggressive orientation, a balanced orientation and a conservative orientation;

[0033] when the credit business orientation is the aggressive orientation, increases with the decrease of the credit risk level l;

[0034] when the credit business orientation is the balanced orientation, does not change with the change of the credit risk level l;

[0035] when the credit business orientation is the conservative orientation, increases with the increase of the credit risk level l.

[0036] In a second aspect, an embodiment of the present application provides an enterprise credit risk determination device, and the device comprises:

[0037] an acquisition module, configured to acquire a plurality of evaluation indexes corresponding to a to-be-evaluated enterprise and feature values corresponding to the evaluation indexes, wherein the plurality of evaluation indexes comprise: an enterprise-dimension evaluation index, an industry-dimension evaluation index and a region-dimension evaluation index;

[0038] a screening module, configured to screen a plurality of target evaluation indexes from the plurality of evaluation indexes based on a combinatorial optimization algorithm;

[0039] a calculation module, configured to determine an enterprise credit risk corresponding to the to-be-evaluated enterprise according to the plurality of target evaluation indexes and the feature values corresponding to the target evaluation indexes.

[0040] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory and a processor.

[0041] The memory stores computer-executed instructions.

[0042] The processor executes the computer-executed instructions stored in the memory, so that the processor executes various possible implementations of any of the above aspects.

[0043] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executed instructions, and the computer-executed instructions are executed by a processor to implement various possible implementations of any of the above aspects.

[0044] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement various possible implementations of any of the above aspects.

[0045] The enterprise credit risk determination method, device, electronic equipment, medium and product provided by the embodiments of the application, the method comprises: acquiring a plurality of evaluation indexes corresponding to an enterprise to be evaluated and feature values corresponding to each evaluation index, the plurality of evaluation indexes comprising: an evaluation index of an enterprise dimension, an evaluation index of an industry dimension, and an evaluation index of a region dimension; based on a combination optimization algorithm, a plurality of target evaluation indexes are selected from the plurality of evaluation indexes; and based on the plurality of target evaluation indexes and the feature values corresponding to each target evaluation index, an enterprise credit risk corresponding to the enterprise to be evaluated is determined. The method can acquire multiple indexes of the enterprise dimension, the industry dimension and the region dimension, not only focuses on the micro financial status of the enterprise itself, but also includes the meso industry prospect (such as the industry cycle and the policy support) and the macro regional environment (such as the regional GDP and the business environment) risk factors, breaks the limitation of the traditional financial index evaluation, includes the forward-looking index, can more comprehensively capture various factors affecting the enterprise credit risk, thereby improving the accuracy of the enterprise credit risk evaluation. And the application can intelligently select the most representative and the most discriminative target evaluation index combination from a large number of initial indexes by the combination optimization algorithm, effectively eliminates the indexes with low credit risk discrimination or high noise, avoids repeated information calculation, further improves the risk identification accuracy, and reduces the risk of misjudgment. BRIEF DESCRIPTION OF DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application.

[0047] Figure 1 An application scenario provided by the embodiments of the application;

[0048] Figure 2 A flowchart of an enterprise credit risk determination method provided by the embodiments of the application;

[0049] Figure 3 A data collection flowchart provided by the embodiments of the application;

[0050] Figure 4 A system architecture diagram provided by the embodiments of the application;

[0051] Figure 5 A structure diagram of an enterprise credit risk determination device provided by the application;

[0052] Figure 6 A structure diagram of an electronic device provided by the application.

[0053] The specific embodiments of the application have been shown by way of example in the above figures, and will be described in greater detail below. These figures and this written description are not intended to limit the scope of the inventive concept in any way, but rather to illustrate the inventive concept to one of ordinary skill in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0054] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The description of the exemplary embodiments is intended to apply to any embodiment of the application, unless specified otherwise. It should be understood that every embodiment need not necessarily include all of the features shown in the drawings or all of the components described in the text, but is intended to include only those features, components or steps that are specifically set forth or otherwise deliberately excluded. Such examples of the exemplary embodiments as are described herein are intended to be illustrative only and are not intended to limit or restrict the scope of the application as claimed in any way.

[0055] In the technical solutions of the present application, the collection, storage, use, processing, transmission, provision and disclosure of information such as financial data or user data comply with relevant laws and regulations and do not violate public order and good customs.

[0056] It should be noted that in the embodiments of the present application, some software, components, models and other existing solutions in the industry may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0057] Enterprise credit risk assessment is the core link of credit decision-making of banks and financial institutions, and the accuracy and scientificity of its results are directly related to the quality and safety of credit assets. An efficient and reliable risk assessment system can effectively identify the credit status of enterprises and accurately distinguish between high-quality customers and high-risk customers, thereby optimizing the allocation of credit resources while controlling bad debt risks, which is of great significance to the stability of the financial system and the promotion of capital services to the real economy.

[0058] At present, the traditional enterprise credit risk assessment in the industry mainly relies on quantitative analysis based on financial indicators. First, a series of financial ratios reflecting the solvency, profitability, operational capacity and development capacity of enterprises are selected as evaluation indicators, such as asset-liability ratio, liquidity ratio, quick ratio, sales profit margin, total asset turnover rate, etc.; then, the experts' experience or through simple statistical methods (such as Analytic Hierarchy Process, AHP) are used to assign certain weights to each indicator; finally, a comprehensive credit score is calculated by using weighted summation or weighted average.

[0059] However, the inventors found through analysis that traditional methods mostly only focus on the financial data and credit history of the enterprise itself, ignoring the influence of external factors such as the industry and regional environment in which the enterprise is located on the credit status of the enterprise, and the traditional methods are mainly based on historical data for static evaluation, which is difficult to make forward-looking predictions on the future credit performance of the enterprise, thereby resulting in low accuracy of enterprise credit risk assessment.

[0060] Therefore, the present application provides an enterprise credit risk determination method, which can obtain a plurality of evaluation indexes corresponding to the enterprise to be evaluated and feature values corresponding to each evaluation index, the plurality of evaluation indexes including evaluation indexes of the enterprise dimension, evaluation indexes of the industry dimension, and evaluation indexes of the regional dimension, and based on a combination optimization algorithm, a plurality of target evaluation indexes are selected from the plurality of evaluation indexes; according to the plurality of target evaluation indexes and the feature values corresponding to each target evaluation index, the enterprise credit risk corresponding to the enterprise to be evaluated is determined. This method can obtain multiple indexes of the enterprise dimension, the industry dimension, and the regional dimension, not only focusing on the micro financial status of the enterprise itself, but also taking into account the medium industry prospects (such as industry cycle, policy support) and the macro regional environment, such as the risk factors of regional GDP (Gross Domestic Product, total production value), breaking the limitations of traditional financial index evaluation, incorporating forward-looking indicators, and being able to more comprehensively capture various factors that affect enterprise credit risk, thereby improving the accuracy of enterprise credit risk assessment. Moreover, the present application can intelligently select the most representative and most discriminant target evaluation index combination from a large number of initial indexes through a combination optimization algorithm, effectively eliminating indexes with low credit risk discrimination or high noise, avoiding repeated information calculation, further improving the risk identification accuracy, and reducing the risk of misjudgment.

[0061] Figure 1 An application scenario diagram provided by an embodiment of the present application is shown in FIG. 1. Figure 1 As shown in FIG. 1, a user inputs a plurality of evaluation indexes corresponding to an enterprise to be evaluated and feature values corresponding to each evaluation index through a client, wherein the plurality of evaluation indexes can include evaluation indexes of the enterprise dimension, evaluation indexes of the industry dimension, and evaluation indexes of the regional dimension. After clicking the confirmation button, the client sends the plurality of evaluation indexes corresponding to the enterprise to be evaluated and the feature values corresponding to each evaluation index to the server, and the server side selects a plurality of target evaluation indexes from the plurality of evaluation indexes based on a combination optimization algorithm. Then, the server determines the enterprise credit risk corresponding to the enterprise to be evaluated according to the plurality of target evaluation indexes and the feature values corresponding to each target evaluation index, and sends the determined enterprise credit risk to the client. The client displays the received enterprise credit risk corresponding to the enterprise to be evaluated on the user interaction interface.

[0062] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0063] Figure 2 A flowchart of an enterprise credit risk determination method provided by an embodiment of the present application is shown in the figure. The execution subject of the embodiment of the present application can be a device with data processing function. The present application takes a server as an example for specific description. As shown in the figure, the enterprise credit risk determination method provided by the embodiment of the present application can include the following steps. Figure 2

[0064] Step 201: Obtain a plurality of evaluation indexes corresponding to a to-be-evaluated enterprise and characteristic values corresponding to the evaluation indexes, wherein the plurality of evaluation indexes include evaluation indexes of enterprise dimensions, evaluation indexes of industry dimensions, and evaluation indexes of regional dimensions.

[0065] The to-be-evaluated enterprise is an enterprise whose credit risk is to be evaluated, and the number of to-be-evaluated enterprises is multiple.

[0066] The plurality of evaluation indexes corresponding to each to-be-evaluated enterprise are the same, but the characteristic values corresponding to the evaluation indexes can be different.

[0067] The evaluation index refers to the attribute or dimension of data, which is a field name. For example, the evaluation index can be debt ratio, liquidity ratio, industry growth rate, and regional GDP. The role of the evaluation index is to define which aspects to observe and describe an enterprise.

[0068] The characteristic value corresponding to the evaluation index refers to the specific numerical value or attribute value of a specific enterprise on the evaluation index. For example, for the evaluation index of debt ratio, the characteristic value of enterprise A is 80%, and the characteristic value of enterprise B can be 60%.

[0069] The evaluation index of enterprise dimensions reflects the micro-quantitative data of the internal operation, management, and financial health status of the to-be-evaluated enterprise, which can directly measure the debt paying ability, profit efficiency, operation level, and development potential of the enterprise individual.

[0070] The evaluation index of industry dimensions reflects the meso-quantitative data of the overall characteristics, development trend, competition pattern, and external environment of the industry where the to-be-evaluated enterprise is located, which measures the systematic risk and opportunity at the industry level.

[0071] The evaluation index of regional dimensions reflects the macro-quantitative data of the economic environment, administrative efficiency, judicial protection, and social culture of the registration place or main operating place of the to-be-evaluated enterprise, which measures the regional risk at the regional level.​

[0072] In an optional implementation, the user inputs or imports, on the client, a plurality of evaluation indexes corresponding to the enterprise to be evaluated and a characteristic value of each evaluation index corresponding to each enterprise to be evaluated, clicks a confirmation button after completion, and the client sends the plurality of evaluation indexes corresponding to the enterprise to be evaluated and the characteristic value of each evaluation index corresponding to each enterprise to be evaluated to the server, and the server acquires the plurality of evaluation indexes corresponding to the enterprise to be evaluated and the characteristic value of each evaluation index corresponding to each enterprise to be evaluated.

[0073] In another optional implementation, the server acquires the plurality of evaluation indexes corresponding to the enterprise to be evaluated and the characteristic value of each evaluation index corresponding to the enterprise to be evaluated through an internal data source and an external data source respectively.

[0074] The internal data source can include enterprise financial statements, credit records, credit reports, etc., and can be acquired through an enterprise or a credit system interface.

[0075] The external data source can include industry statistical data, regional economic data, judicial litigation records, etc., and can be purchased through public channels or data suppliers.

[0076] Figure 3 A data collection process schematic diagram provided by the embodiment of the application is shown in FIG. 1. Figure 3 As shown in FIG. 1, data is collected through an internal data source and an external data source. The collected data is divided into three categories, i.e., structured, semi-structured and unstructured, according to different data formats, and corresponds to different storage carriers.

[0077] Structured data: This type of data has a fixed format and structure, like table data in a relational database, each row and column has a clear definition, so it is stored in a relational database.

[0078] Semi-structured data: The data has a certain structure, but does not strictly follow the fixed table mode like structured data, for example, JSON, XML format data, which is usually stored in a NoSQL database.

[0079] Unstructured data: Data without fixed structure, such as pictures, audio, video, text files, etc. This type of data is stored in a "big data platform".

[0080] After that, the stored structured, semi-structured and unstructured data is preprocessed, through data preprocessing, a structured, clean and reliable data set is formed, laying a foundation for subsequent analysis. Because the original data often has noise, missing, inconsistency and other problems, cleaning and standardization processing is needed first, which includes:

[0081] Noise processing: remove obvious outliers, such as financial indicators beyond reasonable range, etc.

[0082] Missing value processing: select deletion, interpolation, data alignment, etc. according to data characteristics.

[0083] Inconsistent processing: unify evaluation index caliber, such as currency, unit, frequency, etc.

[0084] Data standardization: de-dimensioning of original evaluation indicators, such as Min-Max standardization, Z-score standardization, etc.

[0085] After cleaning and standardization, missing values and outliers are processed, which are common problems of data quality.

[0086] Among them, the missing value processing strategy includes:

[0087] Deletion: if the missing proportion is high, consider deleting the evaluation indicator or sample.

[0088] Mean / Median filling: estimate missing values using the mean or median of the evaluation indicator on other samples.

[0089] KNN interpolation: estimate missing values based on similar sample instances using K-Nearest Neighbor algorithm.

[0090] Matrix filling: treat missing values as matrix completion problem, estimate using low-rank matrix approximation strategy.

[0091] Among them, the outlier processing strategy includes:

[0092] Truncation: replace outliers beyond certain range with boundary values.

[0093] Box-plot detection: detect outliers using quartiles and IQR (Interquartile Range).

[0094] Z-score detection: use mean and standard deviation to determine outliers.

[0095] Isolation forest algorithm: detect outliers by constructing multiple decision trees.

[0096] Optionally, the evaluation indicators of regional dimension include: regional economic development indicators, regional policy support indicators, and regional credit ecological indicators. Regional economic development indicators are used to indicate the overall management situation of the region where the enterprise to be evaluated is located. Regional policy support indicators are used to indicate the credit, industrial policy orientation and strength of the region where the enterprise to be evaluated is located. Regional credit ecological indicators are used to indicate the overall credit ecological situation of the region where the enterprise to be evaluated is located.

[0097] Among them, the regional economic development indicators can include: regional GDP, per capita GDP growth, regional financial revenue, fixed asset investment growth, regional CPI (Consumer Price Index), PPI (Producer Price Index) trend and the like.

[0098] The regional policy support indicators can include: regional credit investment growth, key industry credit proportion, the number and intensity of regional key support industry policy measures.

[0099] The regional credit ecological indicators can include: regional non-performing loan rate, overdue rate level, regional enterprise debt rate, debt pressure index, and regional disposal risk enterprise / project situation.

[0100] In this way, the regional risk is accurately decomposed into three sub-dimensions that are independent of each other and are related to each other: economic foundation, policy orientation, and credit environment, which can accurately identify the source of regional risk. For example, a high-risk enterprise is due to the decline of the local economy (poor economic development indicators), or the policy does not support its industry (poor policy support indicators), or the local generally defaults on payments (poor credit ecological indicators). Such diagnostic capabilities are crucial for risk tracing and developing differentiated risk control strategies, and improve the accuracy of the final enterprise credit risk assessment.

[0101] Step 202, based on a combination optimization algorithm, screening a plurality of target evaluation indicators from a plurality of evaluation indicators.

[0102] Among them, the combination optimization algorithm is an algorithm that finds an optimal solution (maximizes or minimizes a certain objective function value) in a set of a limited number of feasible solutions. The specific form of the combination optimization algorithm is not limited in the present application.

[0103] Based on the combination optimization algorithm, the evaluation indicator combination corresponding to the optimal solution can be determined, and the evaluation indicators in the evaluation indicator combination are all target evaluation indicators.

[0104] Optionally, based on the combination optimization algorithm, a plurality of target evaluation indicators are screened from a plurality of evaluation indicators, including:

[0105] Define decision variables , wherein represents selecting the i-th evaluation indicator, represents not selecting the i-th evaluation indicator;

[0106] Define the objective function , wherein is used to indicate the correlation between the i-th evaluation indicator and the j-th evaluation indicator; wherein n represents the total number of the plurality of evaluation indicators;

[0107] Define constraints , limit the maximum selection of m evaluation indexes;

[0108] Define the total QUBO model , wherein is a preset penalty coefficient;

[0109] Determine the evaluation index combination corresponding to the minimum value of H based on the QAOA algorithm;

[0110] According to the evaluation index combination, determine a plurality of target evaluation indexes.

[0111] Step 1, problem coding, convert the evaluation index selection problem to a QUBO model.

[0112] First, define the decision variable , wherein represents selecting the i-th evaluation index,

[0113] represents not selecting the i-th evaluation index; define the objective function , wherein is used to indicate the correlation between evaluation index i and evaluation index j; define the constraint condition , limit the maximum selection of m evaluation indexes; define the total QUBO model , wherein is a preset penalty coefficient; the goal is to solve the corresponding evaluation index combination.

[0114] Step 2, quantum circuit construction, QAOA (Quantum Approximate Optimization Algorithm) algorithm solves QUBO (Quadratic Unconstrained Binary Optimization) problem by alternately executing two quantum gates:

[0115] Mixing operator acts on the superposition state between the computational basis.

[0116] Phase separation operator , acts to produce more phases at the low value of the objective function.

[0117] Step 3, the performance of QAOA depends on and the selection of the parameters, which need to be optimized by classical optimization algorithms such as gradient descent: initialize Parameters, constructing a QAOA quantum circuit, calculating a loss function , calculating a gradient , updating parameters, repeating steps 2-3 until convergence.

[0118] Step 4, multiple measurements are performed on the optimized QAOA circuit to obtain the optimal solution corresponding to the ranking index combination, which is selected as the multiple target evaluation indexes.

[0119] In this way, the complex business problem of selecting the best index subset is transformed into a rigorous and quantifiable mathematical optimization problem, and the leading quantum computing algorithm (QAOA) is used for efficient solution, improving the efficiency and accuracy of determining the target evaluation index.

[0120] Step 203, determining the enterprise credit risk corresponding to the enterprise to be evaluated according to the multiple target evaluation indexes and the characteristic values corresponding to each target evaluation index.

[0121] Specifically, the server determines the enterprise credit risk corresponding to the enterprise to be evaluated according to the multiple target evaluation indexes corresponding to the enterprise to be evaluated and the characteristic values corresponding to each target evaluation index corresponding to each enterprise to be evaluated.

[0122] The enterprise credit risk determination method provided by the present application can obtain multiple evaluation indexes and characteristic values corresponding to each evaluation index corresponding to the enterprise to be evaluated, the multiple evaluation indexes include: enterprise dimension evaluation indexes, industry dimension evaluation indexes and regional dimension evaluation indexes; based on a combination optimization algorithm, multiple target evaluation indexes are selected from the multiple evaluation indexes; according to the multiple target evaluation indexes and the characteristic values corresponding to each target evaluation index, the enterprise credit risk corresponding to the enterprise to be evaluated is determined, this method can obtain multiple indexes of enterprise dimension, industry dimension and regional dimension, not only focusing on the micro financial status of the enterprise itself, but also including the risk factors of the macro regional environment (such as regional GDP and business environment) and the medium industry prospect (such as industry cycle and policy support), breaking the limitations of traditional financial index evaluation, including forward-looking indexes, which can more comprehensively capture various factors affecting enterprise credit risk, thereby improving the accuracy of enterprise credit risk evaluation, and the present application can intelligently select the most representative and most discriminant target evaluation index combination from a large number of initial indexes through the combination optimization algorithm, effectively eliminating indexes with low credit risk discrimination or high noise, avoiding repeated information calculation, further improving the risk identification accuracy and reducing the risk of misjudgment.

[0123] Optionally, the number of enterprises to be evaluated is multiple, and the enterprise credit risk corresponding to the enterprise to be evaluated is determined according to the multiple target evaluation indexes and the characteristic values corresponding to each target evaluation index, including:

[0124] According to the characteristic values corresponding to each target evaluation index respectively corresponding to each of the plurality of enterprises to be evaluated, a plurality of principal components are extracted from the plurality of target evaluation indexes, and scores corresponding to each principal component respectively corresponding to each of the plurality of enterprises to be evaluated are determined, wherein the proportion of the principal components explaining the variance is greater than a preset threshold value;

[0125] For any enterprise to be evaluated, according to the scores corresponding to each principal component corresponding to the enterprise to be evaluated, the credit comprehensive score corresponding to the enterprise to be evaluated is calculated by using the following formula:

[0126]

[0127] Wherein, The credit comprehensive score corresponding to the pth enterprise to be evaluated is represented by The weight corresponding to the qth principal component is represented by k, and k represents the total number of principal components, The score corresponding to the qth principal component of the pth enterprise to be evaluated is represented by

[0128] Wherein, the weight , wherein The characteristic value corresponding to the qth principal component is represented by

[0129] According to the enterprise credit comprehensive score respectively corresponding to each of the plurality of enterprises to be evaluated, the credit index respectively corresponding to each of the plurality of enterprises to be evaluated is calculated.

[0130] According to the credit index respectively corresponding to each of the plurality of enterprises to be evaluated, the enterprise credit risk corresponding to each of the plurality of enterprises to be evaluated is determined.

[0131] Specifically, the principal components are extracted from the selected target evaluation indexes, and the enterprise credit comprehensive score respectively corresponding to each of the plurality of enterprises to be evaluated is calculated.

[0132] First, the covariance matrix is constructed, and based on the selected target evaluation indexes, the sample covariance matrix is calculated:

[0133]

[0134] Wherein, The characteristic value of the pth enterprise to be evaluated corresponding to the target evaluation index is represented by The average characteristic value of the target evaluation index is represented by c, and c is the total number of enterprises to be evaluated.

[0135] The eigenvalue decomposition is performed on the covariance matrix :

[0136]

[0137] The eigenvalue and the corresponding eigenvector .

[0138] According to the size of the eigenvalue, the first k principal components are selected, and the cumulative variance proportion reaches a preset threshold (such as 80%).

[0139] Based on the principal component score corresponding to the pth to-be-evaluated enterprise Calculate the enterprise credit comprehensive score corresponding to the pth to-be-evaluated enterprise :

[0140]

[0141] Wherein, The credit comprehensive score corresponding to the pth to-be-evaluated enterprise is represented by The weight corresponding to the qth principal component is represented by k, and k represents the total number of principal components, The corresponding score of the qth principal component of the pth to-be-evaluated enterprise is represented by

[0142] Wherein, the weight , wherein The eigenvalue corresponding to the qth principal component is represented by

[0143] Then, the server calculates the credit index corresponding to each to-be-evaluated enterprise according to the enterprise credit comprehensive score corresponding to each to-be-evaluated enterprise, and determines the enterprise credit risk corresponding to each to-be-evaluated enterprise according to the credit index corresponding to each to-be-evaluated enterprise. The credit index and the enterprise credit risk are negatively correlated. The higher the credit index, the lower the enterprise credit risk, and the more the to-be-evaluated enterprise should be prioritized to support the financing demand.

[0144] In this way, through principal component analysis technology, the selected low-redundancy indicators are converted into a set of completely independent principal components, and an objective, stable and strong explanatory credit scoring model is constructed based on this, improving the accuracy of the credit comprehensive score corresponding to each to-be-evaluated enterprise determined, and further improving the accuracy of enterprise credit risk assessment.

[0145] Optionally, the credit index corresponding to each to-be-evaluated enterprise is calculated according to the enterprise credit comprehensive score corresponding to each to-be-evaluated enterprise, including:

[0146] Based on the K-means clustering algorithm, the credit risk grade corresponding to each to-be-evaluated enterprise is determined according to the enterprise credit comprehensive score corresponding to each to-be-evaluated enterprise.

[0147] For any to-be-evaluated enterprise, the credit availability index corresponding to the to-be-evaluated enterprise is determined according to the credit risk grade corresponding to the to-be-evaluated enterprise through the following formula; wherein, The credit availability index corresponding to the pth to-be-evaluated enterprise is represented by indicates the credit risk level corresponding to the to-be-evaluated enterprise; wherein r is the number of the credit risk levels; indicates the weight corresponding to the credit risk level of 1;

[0148]

[0149] According to the credit availability index corresponding to the to-be-evaluated enterprise, the credit scale index corresponding to the to-be-evaluated enterprise is determined by the following formula: used to indicate the scale of the pth to-be-evaluated enterprise;

[0150]

[0151] The credit scale index corresponding to the to-be-evaluated enterprise is determined by the formula The credit index of the to-be-evaluated enterprise is determined by the formula, wherein, is the estimated value of the default loss rate.

[0152] Specifically, the K-means clustering algorithm is used to divide the to-be-evaluated enterprises into different credit risk levels:

[0153] First, the K-means++ algorithm is used to select the initial clustering center:

[0154] Step 1, randomly select a sample as the first clustering center .

[0155] Step 2, for each sample , calculate the shortest distance to the existing clustering center: wherein m refers to the number of clustering centers, indicates the jth clustering center.

[0156] Step 3, select as the new clustering center with a probability of .

[0157] Repeat steps 2-3 until clustering centers are selected.

[0158] Then repeat the following steps until the clustering result converges:

[0159] Step 1, clustering assignment: according to the distance to the clustering center, each sample is assigned to the nearest class:

[0160]

[0161] Step 2, clustering center update: recalculate the center of each class:

[0162]

[0163] wherein c is the number corresponding to the enterprise to be evaluated.

[0164] When the clustering result remains unchanged in two consecutive iterations, or reaches the maximum number of iterations, the algorithm is considered to have converged.

[0165] According to the clustering result, the enterprises are divided into different credit risk levels. The specific division method of the credit risk level is not limited in the present application. Exemplarily, cluster 1: very low credit risk, priority support; cluster 2: low credit risk, moderate support; cluster 3: medium credit risk, cautious treatment; cluster 4: high credit risk, strict control; cluster 5: very high credit risk, avoid credit.

[0166] Optionally, the weight corresponding to each credit risk level is determined according to the credit business orientation, wherein the credit business orientation includes: aggressive orientation, balanced orientation and conservative orientation.

[0167] When the credit business orientation is aggressive orientation, increases with the decrease of the credit risk level l. When the credit business orientation is balanced orientation, does not change with the change of the credit risk level l, that is, the weight corresponding to each credit risk level is relatively balanced.

[0168] When the credit business orientation is conservative orientation, increases with the increase of the credit risk level l.

[0169] In this way, the above method breaks down the barrier between the traditional risk control model and the business strategy, creating an intelligent risk assessment system that can dynamically respond and flexibly adapt to different market environments and bank strategies, making the final risk assessment more in line with the business strategy and improving accuracy.

[0170] After determining the credit risk level corresponding to each enterprise to be evaluated and the weight corresponding to each credit risk level, the server determines the credit availability index corresponding to any enterprise to be evaluated according to the credit risk level corresponding to the enterprise to be evaluated by the following formula; wherein, represents the credit availability index corresponding to the pth enterprise to be evaluated, represents the credit risk level corresponding to the enterprise to be evaluated; wherein r is the number of credit risk levels; represents the weight corresponding to the credit risk level l.

[0171]

[0172] According to the credit availability index corresponding to the enterprise to be evaluated, the credit allocation size index corresponding to the enterprise to be evaluated is determined by the following formula: for indicating the size of the pth enterprise to be evaluated.

[0173]

[0174] The credit allocation size index corresponding to the enterprise to be evaluated is determined by the formula The credit index of the enterprise to be evaluated is determined by the formula , wherein

[0175] The higher the value, the more the financing demand of the enterprise should be prioritized from the perspective of revenue under the premise of controllable credit risk.

[0176] In this way, an automatic, multi-factor weighing mapping chain from the credit comprehensive score to the final credit index is established, not only to satisfy the ranking of enterprises, but also to quantify and integrate the three key dimensions of risk, enterprise size and default loss, and directly output a comprehensive index that can be used to guide credit allocation decisions, greatly improving the accuracy of enterprise credit risk assessment.

[0177] Figure 4 A system architecture diagram is provided for the embodiments of the present application, as shown in Figure 4 The system includes four parts: data layer, algorithm layer, application layer and visualization layer. The data layer is responsible for the collection, cleaning and storage of multi-source heterogeneous data; the algorithm layer includes core algorithm modules such as QAOA optimization, principal component analysis dimension reduction and K-means clustering; the application layer realizes functions such as credit scoring, risk clustering and comprehensive index calculation; and the visualization layer provides a result display and interactive operation interface.

[0178] Corresponding to the enterprise credit risk determination method described above, the embodiments of the present application also provide an enterprise credit risk determination device, Figure 5 A structural diagram of an enterprise credit risk determination device provided by the present application is shown in Figure 5 The enterprise credit risk determination device provided by the present embodiment includes:

[0179] The acquisition module 501 is configured to acquire a plurality of evaluation indexes corresponding to the enterprise to be evaluated and feature values corresponding to each evaluation index, wherein the plurality of evaluation indexes include evaluation indexes of enterprise dimensions, evaluation indexes of industry dimensions and evaluation indexes of regional dimensions.

[0180] The screening module 502 is configured to screen a plurality of target evaluation indexes from the plurality of evaluation indexes based on a combinatorial optimization algorithm.

[0181] The computing module 503 is configured to determine the enterprise credit risk of the enterprise to be evaluated according to the plurality of target evaluation indexes and the characteristic values corresponding to the target evaluation indexes.

[0182] Optionally, the regional dimension evaluation index includes a regional economic development index, a regional policy support index, and a regional credit ecological index. The regional economic development index is used to indicate the overall management situation of the region where the enterprise to be evaluated is located. The regional policy support index is used to indicate the credit and industrial policy orientation and strength of the region where the enterprise to be evaluated is located. The regional credit ecological index is used to indicate the overall credit ecological situation of the region where the enterprise to be evaluated is located.

[0183] Optionally, the screening module 502 is specifically configured to:

[0184] define a decision variable , wherein represents selecting the i th evaluation index, represents not selecting the i th evaluation index;

[0185] define a target function , wherein is used to indicate the correlation between the evaluation index i and the evaluation index j; wherein n represents the total number of the plurality of evaluation indexes;

[0186] define a constraint condition , limit the selection of at most m evaluation indexes;

[0187] define a total QUBO model , wherein is a preset penalty coefficient;

[0188] determine the evaluation index combination corresponding to the minimum value of H based on the QAOA algorithm;

[0189] According to the evaluation index combination, a plurality of target evaluation indexes are determined.

[0190] Optionally, the number of enterprises to be evaluated is a plurality, and the computing module 503 is specifically configured to:

[0191] According to the characteristic values corresponding to the target evaluation indexes corresponding to the plurality of enterprises to be evaluated respectively, a plurality of principal components are extracted from the plurality of target evaluation indexes, and scores corresponding to the principal components corresponding to the plurality of enterprises to be evaluated respectively are determined, wherein the proportion of the principal components explaining the variance is greater than a preset threshold;

[0192] For any enterprise to be evaluated, according to the scores corresponding to the principal components corresponding to the enterprise to be evaluated, the following formula is used to calculate the credit comprehensive score corresponding to the enterprise to be evaluated:

[0193]

[0194] wherein, represents the credit comprehensive score corresponding to the pth enterprise to be evaluated, represents the weight corresponding to the qth principal component, k represents the total number of principal components, represents the corresponding score of the qth principal component of the pth enterprise to be evaluated;

[0195] wherein, the weight wherein represents the eigenvalue corresponding to the qth principal component;

[0196] According to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated, the credit index corresponding to each enterprise to be evaluated is calculated.

[0197] According to the credit index corresponding to each enterprise to be evaluated, the enterprise credit risk corresponding to each enterprise to be evaluated is determined.

[0198] Optionally, when the calculation module 503 calculates the credit index corresponding to each enterprise to be evaluated according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated, it is specifically used for:

[0199] Based on the K-means clustering algorithm, the credit risk level corresponding to each enterprise to be evaluated is determined according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated;

[0200] For any enterprise to be evaluated, the credit availability index corresponding to the enterprise to be evaluated is determined according to the credit risk level corresponding to the enterprise to be evaluated through the following formula; wherein, represents the credit availability index corresponding to the pth enterprise to be evaluated, represents the credit risk level corresponding to the enterprise to be evaluated; wherein r is the number of credit risk levels; represents the weight corresponding to the credit risk level l;

[0201]

[0202] According to the credit availability index corresponding to the enterprise to be evaluated, the credit investment scale index corresponding to the enterprise to be evaluated is determined through the following formula; wherein, is used to indicate the size of the pth enterprise to be evaluated;

[0203]

[0204] The credit investment scale index corresponding to the enterprise to be evaluated is determined through the formula , wherein, is the estimated value of the default loss rate.

[0205] Optionally, is determined according to a credit business orientation, wherein the credit business orientation comprises: an aggressive orientation, a balanced orientation and a conservative orientation;

[0206] when the credit business orientation is the aggressive orientation, increases with the decrease of the credit risk level l;

[0207] when the credit business orientation is the balanced orientation, does not change with the change of the credit risk level l;

[0208] when the credit business orientation is the conservative orientation, increases with the increase of the credit risk level l.

[0209] The enterprise credit risk determination apparatus provided in the embodiment can execute the method provided in the method embodiment, and has similar implementation principles and technical effects, which will not be described here.

[0210] Figure 6 The structure of the electronic device provided in the present application is shown in the figure. Figure 6 As shown in the figure, the electronic device 60 provided in the embodiment comprises at least one processor 601 and a memory 602. Optionally, the device 60 further comprises a communication component 603. The processor 601, the memory 602 and the communication component 603 are connected through a bus 604.

[0211] In the specific implementation process, the at least one processor 601 executes the computer execution instructions stored in the memory 602, so that the at least one processor 601 executes the above-mentioned method.

[0212] The specific implementation process of the processor 601 can refer to the method embodiment, which has similar implementation principles and technical effects, and will not be described here.

[0213] In the above-mentioned embodiment, it should be understood that the processor can be a central processing unit (English: Central Processing Unit, CPU for short), and can also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, DSP for short), application specific integrated circuits (English: Application Specific Integrated Circuit, ASIC for short) and the like. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor, etc. The steps of the method disclosed in the present application can be directly embodied as the execution of the hardware processor, or executed by the combination of hardware and software modules in the processor.

[0214] The memory can include a Random Access Memory (RAM) and can also include a Non-volatile Memory (NVM), such as at least one disk memory.

[0215] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0216] The present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the above method.

[0217] The present application also provides a computer readable storage medium, which stores computer execution instructions, and when a processor executes the computer execution instructions, the above method is implemented.

[0218] The above readable storage medium can be implemented by any type of volatile or non-volatile storage device or their combination, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0219] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0220] The division of units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0221] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, can be located in one place, or can be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0222] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0223] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0224] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. The program executes the steps including the above-mentioned method embodiments when executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various program code storage media.

[0225] It should be understood that many of the materials and devices exemplified in this disclosure are articles of manufacture (i.e., articles of manufacture) according to this disclosure. The articles of manufacture can be manufactured as such or can be manufactured by combining the materials and devices exemplified in this disclosure. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. It should be understood that, in some embodiments, equivalents to the specific electrode structures and / or methods described herein can be employed without departing from the scope of the application. Accordingly, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," "having," "containing," "involving," "characterized by," "characterized into," and variations thereof herein, is meant to encompass the items listed thereafter, and equivalents thereof as well as additional items. Although the foregoing application has been described in some detail by way of illustration and example, it is not to be limited thereby, but rather, only by the scope of the appended claims.

Claims

1. A method of determining business credit risk, characterized by, The method comprises: obtaining a plurality of evaluation indexes corresponding to an enterprise to be evaluated and characteristic values corresponding to each evaluation index, wherein the plurality of evaluation indexes comprise evaluation indexes of enterprise dimensions, evaluation indexes of industry dimensions and evaluation indexes of regional dimensions; based on a combinatorial optimization algorithm, a plurality of target evaluation indexes are selected from the plurality of evaluation indexes; determine the enterprise credit risk of the enterprise to be evaluated according to the plurality of target evaluation indexes and the characteristic values corresponding to each target evaluation index.

2. The method of claim 1, wherein, The evaluation indexes of the regional dimensions comprise regional economic development indexes, regional policy support indexes and regional credit ecological indexes, the regional economic development indexes are used to indicate the overall management situation of the region where the enterprise to be evaluated is located, the regional policy support indexes are used to indicate the credit, industrial policy orientation and strength situation of the region where the enterprise to be evaluated is located, and the regional credit ecological indexes are used to indicate the overall credit ecological situation of the region where the enterprise to be evaluated is located.

3. The method of claim 1, wherein, Based on the combinatorial optimization algorithm, a plurality of target evaluation indexes are selected from the plurality of evaluation indexes, comprising: Defining decision variables wherein denotes the selection of the i-th evaluation criterion, denotes the non-selection of the i-th evaluation criterion; Defining an objective function wherein for indicating the correlation of evaluation index i and evaluation index j; wherein n represents the total number of the plurality of evaluation indexes; Define constraints , limit the maximum number of selected evaluation indicators m; Defining the overall QUBO model wherein is a preset penalty coefficient; determining the evaluation index combination corresponding to the minimum value of H based on the QAOA algorithm; determine a plurality of target evaluation indexes according to the evaluation index combination.

4. The method according to any one of claims 1 to 3, characterized in that, The number of the enterprises to be evaluated is a plurality, and the enterprise credit risk of the enterprise to be evaluated is determined according to the plurality of target evaluation indexes and the characteristic values corresponding to each target evaluation index, comprising: extract a plurality of principal components from the plurality of target evaluation indexes according to the characteristic values corresponding to each target evaluation index corresponding to a plurality of enterprises to be evaluated respectively, and determine the scores corresponding to each principal component corresponding to a plurality of enterprises to be evaluated respectively, wherein the proportion of the principal components explaining the variance is greater than a preset threshold; for any enterprise to be evaluated, the credit comprehensive score corresponding to the enterprise to be evaluated is calculated according to the scores corresponding to each principal component corresponding to the enterprise to be evaluated by using the following formula: wherein, represents the credit comprehensive score corresponding to the pth enterprise to be evaluated, represents the weight corresponding to the qth principal component, k represents the total number of principal components, represents the corresponding score of the qth principal component of the pth enterprise to be evaluated; wherein the weight wherein denotes the eigenvalue corresponding to the qth principal component; calculate the credit index corresponding to each enterprise to be evaluated according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated respectively; determine the enterprise credit risk corresponding to each enterprise to be evaluated according to the credit index corresponding to each enterprise to be evaluated respectively.

5. The method of claim 4, wherein, According to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated respectively, the credit index corresponding to each enterprise to be evaluated respectively is calculated, comprising: based on the K-means clustering algorithm, the credit risk level corresponding to each enterprise to be evaluated respectively is determined according to the enterprise credit comprehensive score corresponding to each enterprise to be evaluated respectively. For any to-be-evaluated enterprise, according to the credit risk grade corresponding to the to-be-evaluated enterprise, the credit availability index corresponding to the to-be-evaluated enterprise is determined through the following formula; wherein, represents the credit availability index corresponding to the pth to-be-evaluated enterprise, represents the credit risk grade corresponding to the to-be-evaluated enterprise; wherein r is the number of the credit risk grades; represents the weight corresponding to the credit risk grade l; According to the credit availability index corresponding to the enterprise to be evaluated, the credit allocation size index corresponding to the enterprise to be evaluated is determined by the following formula; wherein, is used to indicate the size of the pth enterprise to be evaluated; The credit investment scale index corresponding to the enterprise to be evaluated is determined by the formula , wherein is an estimated value of the default loss rate.

6. The method of claim 5, wherein, is determined according to a credit business orientation, wherein the credit business orientation comprises: an aggressive orientation, a balanced orientation and a conservative orientation; When the credit business orientation is aggressive, increases with decreasing credit risk class l; In the credit business orientation is balanced, does not change with the credit risk level l; When the credit business orientation is conservative, increases with the credit risk class l.

7. An enterprise credit risk determination apparatus characterized by comprising: The device comprises: an acquisition module for acquiring a plurality of evaluation indexes corresponding to an enterprise to be evaluated and characteristic values corresponding to each evaluation index, wherein the plurality of evaluation indexes comprise evaluation indexes of enterprise dimensions, evaluation indexes of industry dimensions and evaluation indexes of regional dimensions; a screening module for selecting a plurality of target evaluation indexes from the plurality of evaluation indexes based on a combinatorial optimization algorithm; a calculation module for determining the enterprise credit risk of the enterprise to be evaluated according to the plurality of target evaluation indexes and the characteristic values corresponding to each target evaluation index.

8. An electronic device, comprising: comprise: a memory, a processor; the memory stores computer execution instructions; The processor executes computer-executable instructions stored in the memory such that the processor performs the method of any of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium has stored therein computer-executable instructions that, when executed by a processor, perform the method of any of claims 1-6.

10. A computer program product, characterised in that, A computer program that, when executed by a processor, performs the method of any of claims 1-6.

Images