User credit analysis method and device, equipment and storage medium

Abstract

The invention provides a user credit analysis method and device, equipment and a storage medium. The method comprises the steps of obtaining first data of a target influence factor of a to-be-analyzed object, and then inputting the first data of the target influence factor into a pre-constructed cross decision table to obtain a credit score corresponding to the to-be-analyzed object; the cross decision table is determined based on the historical data of the multiple influence factors and the incidence relation among the multiple influence factors, and finally determining the credit analysis result of the to-be-analyzed object according to the credit score. According to the method provided by the invention, the credit analysis of the user is realized, the reliability of the analysis process is improved, and the overfitting of the decision-making process is reduced.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a user credit analysis method, device, equipment and storage medium. Background Art

[0002] For credit rating, we first need to identify and classify the key factors of existing credit ratings, such as enterprise size, financial status, profitability, operating stability, industry prospects, etc., classify these factors, and for each rating factor, use the specific values ​​of financial indicators, such as main business income, accounts receivable, accounts payable and other quantitative data, combined with industry outlook indicators (such as optimism, pessimism, neutrality and other qualitative factors) to analyze and formulate corresponding evaluation standards.

[0003] Existing technologies mainly implement credit rating rules by using decision trees, which are defined according to the appropriate split points of the distribution of each characteristic factor, and the decision-making process and results are displayed in the form of a tree diagram.

[0004] However, in the prior art, the decision tree ignores the correlation between key factors, and has the problems of poor reliability in the decision-making process and prone to overfitting or underfitting. Summary of the invention

[0005] The present application provides a user credit analysis method, apparatus, device and storage medium to solve the technical problems of poor reliability and easy overfitting or underfitting in the decision-making process of user credit analysis.

[0006] In a first aspect, the present application provides a user credit analysis method, comprising:

[0007] Acquire first data of a target influencing factor of the object to be analyzed;

[0008] Inputting the first data of the target influencing factor into a pre-constructed cross decision table to obtain a credit score corresponding to the object to be analyzed, wherein the cross decision table is determined based on historical data of multiple influencing factors and correlations between the multiple influencing factors;

[0009] A credit analysis result of the object to be analyzed is determined according to the credit score.

[0010] In one or more embodiments, before inputting the first data of the target influencing factor into a pre-built cross decision table to obtain the credit score corresponding to the object to be analyzed, the method further includes:

[0011] Structurally processing the historical data of the multiple influencing factors to obtain at least one two-dimensional format data;

[0012] Performing association processing on the at least one two-dimensional format data to obtain a key factor;

[0013] configuring decision conditions for the key factors and determining decision standard information;

[0014] The cross decision table is constructed according to the key factors, the decision conditions corresponding to the key factors, and the decision standard information.

[0015] In one or more embodiments, the performing association processing on the at least one two-dimensional format data to obtain the key factor includes:

[0016] The at least one two-dimensional format data is analyzed by using a preset tool to obtain influencing factors with a correlation greater than a first preset threshold value, which are recorded as the key factors.

[0017] In one or more embodiments, the performing association processing on the at least one two-dimensional format data to obtain the key factor includes:

[0018] Performing statistics on the at least one two-dimensional format data to determine the correlation coefficient between various influencing factors;

[0019] The influencing factors whose correlation coefficients are greater than the second preset threshold are recorded as the key factors.

[0020] In one or more embodiments, determining the decision criterion information includes:

[0021] configuring weight coefficients for the key factors;

[0022] The decision criterion information is determined according to the weight coefficient.

[0023] In one or more embodiments, the method further comprises:

[0024] S1, testing the cross decision table to obtain a test result;

[0025] S2, optimizing the cross decision table according to the test result to obtain an optimized cross decision table.

[0026] In one or more embodiments, the optimizing the cross decision table according to the test result to obtain an optimized cross decision table includes:

[0027] When the test result is lower than the first threshold, the cross decision table is adjusted, and the above steps S1-S2 are repeated until the test result is greater than or equal to the first threshold. The obtained cross decision table is recorded as an optimized cross decision table.

[0028] In a second aspect, the present application provides a user credit analysis device, comprising:

[0029] An acquisition module, used to acquire first data of a target influencing factor of the object to be analyzed;

[0030] A processing module, configured to input the first data of the target influencing factor into a pre-constructed cross-decision table to obtain a credit score corresponding to the object to be analyzed, wherein the cross-decision table is determined based on historical data of multiple influencing factors and correlations between the multiple influencing factors;

[0031] A determination module is used to determine the credit analysis result of the object to be analyzed according to the credit score.

[0032] In one embodiment, before inputting the first data of the target influencing factor into a pre-built cross decision table to obtain the credit score corresponding to the object to be analyzed, the processing module is further used to:

[0033] Structurally processing the historical data of the multiple influencing factors to obtain at least one two-dimensional format data;

[0034] Performing association processing on the at least one two-dimensional format data to obtain a key factor;

[0035] configuring decision conditions for the key factors and determining decision standard information;

[0036] The cross decision table is constructed according to the key factors, the decision conditions corresponding to the key factors, and the decision standard information.

[0037] In one embodiment, the processing module performs association processing on the at least one two-dimensional format data to obtain the key factor, specifically for:

[0038] The at least one two-dimensional format data is analyzed by using a preset tool to obtain influencing factors with a correlation greater than a first preset threshold value, which are recorded as the key factors.

[0039] In one embodiment, the processing module performs association processing on the at least one two-dimensional format data to obtain the key factor, specifically for:

[0040] Performing statistics on the at least one two-dimensional format data to determine the correlation coefficient between various influencing factors;

[0041] The influencing factors whose correlation coefficients are greater than the second preset threshold are recorded as the key factors.

[0042] In one embodiment, the processing module, the determining decision criterion information, is specifically used to:

[0043] configuring weight coefficients for the key factors;

[0044] The decision criterion information is determined according to the weight coefficient.

[0045] In one embodiment, the processing module is further used for:

[0046] S1, testing the cross decision table to obtain a test result;

[0047] S2, optimizing the cross decision table according to the test result to obtain an optimized cross decision table.

[0048] In one embodiment, the processing module optimizes the cross decision table according to the test result to obtain an optimized cross decision table, which is specifically used to:

[0049] When the test result is lower than the first threshold, the cross decision table is adjusted, and the above steps S1-S2 are repeated until the test result is greater than or equal to the first threshold. The obtained cross decision table is recorded as an optimized cross decision table.

[0050] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0051] The memory stores computer-executable instructions;

[0052] The processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect and any one of the embodiments above.

[0053] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect and any one of the embodiments.

[0054] The user credit analysis method, device, equipment and storage medium provided by the present application relate to the field of data analysis technology. The method obtains the first data of the target influencing factors of the object to be analyzed, and then inputs the first data of the target influencing factors into a pre-constructed cross-decision table to obtain the credit score corresponding to the object to be analyzed. The cross-decision table is determined based on the historical data of multiple influencing factors and the correlation between the multiple influencing factors. Finally, the credit analysis result of the object to be analyzed is determined according to the credit score. In the above method, the accuracy and efficiency of the decision-making process are improved by pre-constructing a cross-decision table. Based on the decision-making influencing factors and decision conditions with correlation relationships in the constructed cross-decision table, the first data of the target influencing factors of the object to be analyzed is input into the pre-constructed cross-decision table, and the credit of the object to be analyzed is analyzed, which effectively improves the reliability of the analysis process and reduces the overfitting of the decision-making process. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0056] Figure 1 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 1 ;

[0057] Figure 2 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 2 ;

[0058] Figure 3 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 3 ;

[0059] Figure 4 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 4 ;

[0060] Figure 5 A schematic diagram of the structure of a user credit analysis device provided in an embodiment of the present application;

[0061] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0062] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0063] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0064] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0065] Credit rating is mainly to evaluate the credit of users comprehensively and objectively, so as to provide valuable user credit information for relevant enterprises or industries. For credit rating, it is necessary to first identify and classify the key factors of existing credit ratings, such as enterprise size, financial status, industry prospects, etc. For each key factor of credit rating, use the specific values ​​of financial indicators, such as main business income, accounts receivable and payable, and other quantitative data, combined with industry prospect indicators for specific analysis, formulate corresponding evaluation standards, and make decisions based on the results of credit evaluation.

[0066] Existing technologies mainly implement credit rating rules through decision trees, and display the decision-making process and results in the form of a tree data structure. It is defined according to the appropriate split point of each characteristic factor distribution. Each internal node represents the judgment of a key factor, each branch represents the output of a judgment result, and finally each leaf node represents a decision result.

[0067] However, in the prior art, the decision tree ignores the correlation between key factors, and has the problems of poor reliability and prone to overfitting or underfitting.

[0068] The method, device and storage medium for user credit analysis provided by the present application are intended to solve the above technical problems of the prior art. The inventive concept of the present application is as follows: by obtaining the first data of the target influencing factors of the object to be analyzed and inputting it into a pre-constructed cross-decision table, the credit score of the object to be analyzed is obtained. By pre-constructing the cross-decision table, the reliability of the analysis process can be improved and the problem of overfitting can be avoided. When pre-constructing the cross-decision table, the historical data of multiple influencing factors and the correlation between multiple influencing factors are determined to ensure the correlation between the influencing factors. Finally, according to the credit score, the credit analysis result of the object to be analyzed is determined, and the credit analysis of the user is completed.

[0069] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are 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 may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0070] Figure 1 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 1 .like Figure 1 As shown, the execution subject is an electronic device that implements the user credit analysis method, which includes the following steps:

[0071] S110: Acquire first data of a target influencing factor of the object to be analyzed.

[0072] The object to be analyzed may be a user or an enterprise, and the influencing factors include various types of data related to user credit analysis, such as age, education, income, expenditure, etc. The first data is the specific content of various data types, such as an age value of twenty and an annual income of two hundred thousand.

[0073] In this step, the first data obtained should reduce missing values ​​and null values ​​as much as possible, identify and process duplications, errors, inconsistencies, etc., so that the data has a unified data format, unit, encoding, etc. Cleaning and preprocessing the first data include: missing value processing, outlier processing, data type conversion, and deduplication.

[0074] Missing value processing includes: directly deleting the data item record when there are many missing data items, estimating a possible value based on related data items to fill in the missing data item;

[0075] Outlier processing includes: directly deleting data that exceeds the norm, such as age exceeding 200. For other abnormal situations, outliers can be replaced with medians, etc.

[0076] Data type conversion: The data is processed in a unified manner, for example, dates are in the format of yyyy-MM-dd hh:mi:ss, and encoding is in utf8 mode;

[0077] Deduplication: Use the pivot table to summarize and deduplicate data by default, identify duplicate data records, retain one data record, and delete other duplicates.

[0078] S120: Input the first data of the target influencing factor into a pre-constructed cross decision table to obtain a credit score corresponding to the object to be analyzed. The cross decision table is determined based on historical data of multiple influencing factors and correlations between the multiple influencing factors.

[0079] The pre-built cross decision table includes decision rows and decision columns. The historical data of multiple influencing factors are input into the cross decision table to determine the influencing factors corresponding to the decision rows and decision columns, and there is a correlation between the multiple influencing factors.

[0080] In this step, the first data of the target influencing factor is input into a pre-constructed cross decision table, and the specific content in the first data corresponding to the decision row and decision column influencing factors of the pre-constructed cross decision table is determined, and then a specific decision is made according to the decision conditions of the pre-constructed cross decision table to obtain the credit score corresponding to the object to be analyzed.

[0081] S130. Determine the credit analysis result of the object to be analyzed according to the credit score.

[0082] The credit score may be a specific number within a certain value range, which is used to feedback the credit analysis result of the object to be analyzed, and the credit analysis result may be determined according to different values ​​of the credit score.

[0083] For example, if the credit score is an integer between 0 and 100, the higher the score, the better the credit of the analyzed object. When the credit score is between 0 and 20, the corresponding credit analysis result is D, that is, "the credit of the analyzed object is poor"; when the credit score is between 21 and 40, the corresponding credit analysis result is C, that is, "the credit of the analyzed object is relatively poor"; when the credit score is between 41 and 60, the corresponding credit analysis result is B, that is, "the credit of the analyzed object is average"; when the credit score is between 61 and 80, the corresponding credit analysis result is A, that is, "the credit of the analyzed object is good"; when the credit score is between 81 and 100, the corresponding credit analysis result is S, that is, "the credit of the analyzed object is very good".

[0084] The user credit analysis method provided in the embodiment of the present application obtains the first data of the target influencing factors of the object to be analyzed, inputs the first data into a pre-constructed cross-decision table, obtains the credit score corresponding to the object to be analyzed, and then determines the credit analysis result of the object to be analyzed. In this embodiment, the first data of the target influencing factors of the object to be analyzed is pre-processed, and duplicate and erroneous data are identified and processed, so that the data has a unified data format, which improves the efficiency and accuracy of data analysis, and by pre-constructing a cross-decision table, the credit of the object to be analyzed is analyzed based on the decision influencing factors and decision conditions with associated relationships in the constructed cross-decision table, which effectively improves the reliability of the analysis process and reduces the overfitting of the decision process.

[0085] Based on the above embodiments, Figure 2 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 2 .like Figure 2 As shown, before step S120, the method may further include the following implementation:

[0086] S210 , structurally process the historical data of multiple influencing factors to obtain at least one two-dimensional format data.

[0087] Among them, structured processing refers to organizing the historical data of multiple influencing factors into a structured format, sorting and storing them as two-dimensional table data, where each column of the table represents an influencing factor, and each row represents a historical data record of the object to be processed.

[0088] For example, the multiple influencing factors of the object to be processed include age, education level, monthly income, etc., then these influencing factors are used as attributes of each column of the two-dimensional table, and the historical data of the multiple influencing factors of each object to be processed are used as the value of each row of the two-dimensional table. For example, the historical data of object X1 is {35 years old, Y1 education level, 10000}, the historical data of object X2 is {40 years old, Y2 education level, 15000}, the historical data of object X3 is {26 years old, Y2 education level, 12000}, and the historical data of object X4 is {31 years old, Y1 education level, 11000}. In this way, a two-dimensional table is constructed and the historical data of multiple influencing factors are stored. The influencing factors in the first column are age, and the values ​​are {35 years old, 40 years old, 26 years old, 31 years old} in sequence, the influencing factors in the second column are education level, and the values ​​are {Y1 education level, Y2 education level, Y2 education level, Y1 education level} in sequence, and the influencing factors in the third column are monthly income, and the values ​​are {10000, 15000, 12000, 11000} in sequence.

[0089] S220: Perform association processing on at least one two-dimensional format data to obtain key factors.

[0090] Among them, association processing refers to analyzing the correlation between historical data of multiple influencing factors based on two-dimensional format data, and analyzing and obtaining relevant influencing factors with correlation according to the specific quantitative relationship between data values, and taking these influencing factors as key factors.

[0091] In a possible implementation, S220 may include: using a preset tool to analyze at least one two-dimensional format data, obtaining an influencing factor with a correlation greater than a first preset threshold, and recording it as a key factor.

[0092] Among them, the preset tool can be a Chatgpt tool, and the first preset threshold can be an artificially set correlation judgment standard. For example, the correlation value range is 0 to 1. The closer the correlation value is to 1, the greater the correlation between the influencing factors. The first preset threshold can be 0.7. When the correlation value is greater than or equal to 0.7, it means that there is a correlation relationship between the influencing factors, and these influencing factors are recorded as key factors.

[0093] In this step, the two-dimensional format data needs to be integrated into a data format that can be accessed by the Chatgpt tool, that is, the data is converted into structured or semi-structured data in Excel-related software. It does not involve text conversion, but only fills the data into specified rows or columns. First, use Structured Query Language (SQL) or code to desensitize some sensitive data to avoid leaking user detailed data on the Internet. Then upload the cleaned data to Chatgpt through Chrome's extension tool ChatGPT File Uploader Extended. After the data is uploaded, you need to enter instructions in the Chatgpt interface to indicate that the uploaded data needs to be cross-analyzed. For example, enter: "Please analyze the cross-relationship between age and education in this data set, find out the mathematical relationship between age and education, and give a detailed mathematical formula." After receiving the request, Chatgpt cross-analyzes the data and generates results. It verifies whether the mathematical formula output by Chatgpt is correct through relevant historical data. If not, enter more keywords in the context to make sure Chatgpt can understand the requirements. Then, iterate and optimize the generated results by continuously entering Chatgpt commands, and finally obtain the cross-relationship between age and education in the data set to reflect the correlation between age and education.

[0094] In a possible implementation, S220 may include: performing statistics on at least one two-dimensional format data to determine the correlation coefficient between various influencing factors; and recording the influencing factors with correlation coefficients greater than a second preset threshold as key factors.

[0095] Among them, the correlation coefficient can be the Pearson correlation coefficient or the Spearman rank correlation coefficient, and the second preset threshold can be an artificially set correlation judgment standard. For example, the value range of the Pearson correlation coefficient is -1 to 1. The closer the value of the correlation coefficient is to 1, the greater the correlation between the influencing factors. The second preset threshold can be 0.8. When the value of the correlation coefficient is greater than or equal to 0.8, it means that the influencing factors are correlated, and these influencing factors are recorded as key factors.

[0096] In this step, first, the historical data of the influencing factors (such as education, age, and income) are plotted into a scatter plot, and the relationship between these historical data is observed. Then, the correlation coefficient is calculated using statistical software (such as SPSS, Excel, R, Python, etc.), and the Pearson correlation coefficient is used to measure the linear relationship, and the Spearman rank correlation coefficient is used to measure the monotonic relationship. After that, the strength and direction of the relationship between the influencing factors are determined based on the calculated correlation coefficient. For example, the value range of the Pearson correlation coefficient is -1 to 1, where 0 indicates no correlation, a positive value indicates a positive correlation, and a negative value indicates a negative correlation. Find the data with positive correlation values ​​close to 1, and use hypothesis testing to determine whether the data impact of these influencing factors with associated relationships is significant. That is, after finding the data of influencing factors with associated relationships, select independent variables and dependent variables, and select appropriate regression models according to the nature of the data and the purpose of the research. Use statistical software to fit the data, estimate the parameters of the fitting model formula, bring the independent variables into the model formula, and obtain the corresponding dependent variable values ​​based on the values ​​of the independent variables. Compare them with the provided result set to verify the importance of the influencing factor. If the calculated dependent variable value is the same as or has a small difference with the value corresponding to the provided result set, it means that the influencing factor corresponding to the dependent variable is more important.

[0097] S230. Configure decision conditions for key factors and determine decision standard information.

[0098] Among them, the decision conditions can be based on key factors, and different value ranges can be configured for different key factors. The decision standard information can be determined according to the weighted average method, and the decision standard information can reflect the user's credit status.

[0099] In a possible implementation, S230 may include: configuring weight coefficients for key factors, and determining decision criterion information according to the weight coefficients.

[0100] Among them, the weight coefficient can be configured according to the importance of the key factor to the user's credit, and the value range is 0 to 1.

[0101] For example, the basic score M is obtained from the user's financial income and operating ability in the key factors, the extended score N is obtained from the user's financial expenditure in the key factors, and the credit score O is obtained from the user's historical transaction record performance in the key factors. The weight of the basic score M is configured to 0.5, the weight of the extended score N is configured to 0.2, and the weight of the credit score O is configured to 0.3. Finally, the weighted average score V is calculated according to the formula M*0.5+N*0.2+O*0.3. The weighted average V is used to reflect the decision standard information. Among them, the value ranges of the basic score M, the extended score N and the credit score O are all set to 0 to 10, and the corresponding weighted average score V has a numerical range of 0 to 10. When the weighted average score V is 0 to 4, it means "the user's credit is average", when the weighted average score V is 4 to 8, it means "the user's credit is good", and when the weighted average score V is 8 to 10, it means "the user's credit is very good".

[0102] S240: construct a cross decision table according to the key factors, the decision conditions corresponding to the key factors, and the decision standard information.

[0103] In this step, some key factors are used as decision rows of the cross decision table, and other key factors are used as decision columns of the cross decision table. The decision conditions corresponding to the key factors are set, and the obtained decision standard information is filled into the cross decision table to complete the construction of the cross decision table.

[0104] Taking the above example for further explanation, the user's financial income and management ability are used as the decision rows of the cross decision table, and the decision conditions of the user's financial income are set as "annual income greater than 500,000", "annual income less than 500,000 and greater than 200,000", and "annual income less than 200,000". The decision conditions of the user's management ability are set as "strong management ability" and "weak management ability". When the user's annual income is greater than 500,000 and the management ability is strong, the basic score A is 10 points. When the user's annual income is greater than 500,000 and the management ability is weak, the basic score A is 8 points. When the user's annual income is less than 500,000 and greater than 200,000 and the management ability is strong, the basic score A is 6 points. When the user's annual income is less than 500,000 and greater than 200,000 and the management ability is weak, the basic score A is 4 points. When the user's annual income is less than 200,000 and the management ability is strong, the basic score A is 2 points. When the user's annual income is less than 200,000 and the management ability is weak, the basic score A is 0 points. The user's financial expenditure and the user's historical transaction record performance are used as the decision columns of the cross decision table. The decision condition of the user's financial expenditure is set to "annual expenditure greater than 300,000", and the corresponding extended score N is 10 points. The corresponding extended score N is 5 points for "annual expenditure less than 300,000 and greater than 100,000", and the corresponding extended score N is 3 points for "annual expenditure less than 100,000". The decision conditions of the user's historical transaction record performance are set to "excellent", "good", and "average", and the corresponding credit scores O are 10 points, 6 points, and 3 points respectively. The weighted average score V is calculated according to the formula M*0.5+N*0.2+O*0.3. When the user's annual income is less than 500,000 and greater than 200,000 and his / her management ability is weak, and his / her annual expenditure is less than 300,000 and greater than 100,000, and his / her historical transaction record is good, the weighted average score V is 4.8 points, that is, the corresponding decision standard information is 4.8 points. The result is filled in the cross-decision table unit defined by the key factors and the decision conditions corresponding to the key factors, that is, the decision behavior is "the user's annual income is less than 500,000 and greater than 200,000" and "the management ability is weak", and the decision is listed as "the annual expenditure is less than 300,000 and greater than 100,000" and "the historical transaction record is good".

[0105] The user credit analysis method provided in the embodiment of the present application performs structured processing on the historical data of multiple influencing factors to obtain at least one two-dimensional format data, and performs association processing on the two-dimensional format data to obtain key factors, then configures decision conditions for the key factors, and determines decision standard information, and finally constructs a cross-decision table based on the obtained key factors, the decision conditions corresponding to the key factors, and the decision standard information. In this embodiment, by performing structured processing on the historical data of multiple influencing factors, the data is organized into two-dimensional structured data, which facilitates the storage and analysis of the data, and then the two-dimensional format data is associated, the correlation between multiple influencing factors is analyzed, and the influencing factors with strong correlation are selected as key factors, which improves the reliability of the decision, and the cross-decision table is constructed based on the obtained key factors, the decision conditions corresponding to the key factors, and the decision standard information, which effectively avoids overfitting in the analysis process.

[0106] Based on the above embodiments, Figure 3 Schematic diagram of the process of the user credit analysis method provided in the embodiment of the present application Figure 3 .like Figure 3 As shown, the method also includes the following steps:

[0107] S1, test the cross decision table and obtain the test results.

[0108] The test refers to comparing the score of a certain evaluation subject with a predetermined score index, and the test result can be the similarity of the comparison.

[0109] In this step, the accuracy and reliability of the cross decision table are tested by using a validation data set, which is taken from existing historical data and corresponding score indicators. The data of multiple influencing factors of the validation data set are input into the cross decision table, and the score of the evaluation subject is obtained according to the decision standard information of the cross decision table. The score of the evaluation subject is compared with the score indicator corresponding to the historical data in the validation data set to obtain the similarity between the score of the evaluation subject and the score indicator corresponding to the historical data.

[0110] S2, optimizing the cross decision table according to the test results to obtain an optimized cross decision table.

[0111] In this step, the optimization processing can be to use an optimization method to optimize the key factors of the cross decision table, for example, a genetic algorithm is used to map the key factors in the cross decision table, such as age, salary, education and other related information, into real number vectors, and then randomly assign these vectors to different groups, input the data of the key factors included in different groups into the actual cross decision table for trial, find the group with higher fitness, that is, the group with higher decision standard information value, and then select the key factors of the group with higher fitness for random combination, form a new group with some of the key factors with better performance, observe the decision standard information of the new group in the cross decision table, and re-select the key factors with better performance to become an excellent group, and finally reach the optimal key factor through continuous iteration, and use the optimal key factor as the new key factor in the cross decision table.

[0112] In a possible implementation, S2 may include: when the test result is lower than the first threshold, adjusting the cross decision table, and repeating the above steps S1-S2 until the test result is greater than or equal to the first threshold, and the obtained cross decision table is recorded as an optimized cross decision table.

[0113] In this step, the first threshold value may be a pre-set threshold indicator for reflecting the accuracy and reliability of the cross decision table. If the first threshold value is a preset similarity threshold value, when the similarity is large and greater than the threshold value, it indicates that the accuracy and reliability of the cross decision table are good. When the similarity is small and less than the threshold value, it indicates that the accuracy and reliability of the cross decision table are poor. For example, the numerical range of the similarity is 0% to 100%, and the similarity threshold value, i.e., the first threshold value, is set to 80%. When the similarity is less than 80%, the setting of the cross decision table is adjusted, and steps S1-S2 are repeated until the similarity is greater than or equal to 80%. The cross decision table obtained at this time is recorded as the optimized cross decision table.

[0114] The user credit analysis method provided in the embodiment of the present application tests the cross decision table to obtain the test results, and then optimizes the cross decision table according to the test results to finally obtain the optimized cross decision table. In this embodiment, the cross decision table is tested, and the test results obtained reflect the accuracy and reliability of the constructed cross decision table. Then, according to the test results, the cross decision table is optimized to improve the accuracy and reliability of the cross decision table. Finally, the optimized cross decision table has high accuracy and reliability, and can effectively avoid overfitting in the analysis process.

[0115] Based on the above embodiments, Figure 4 The process of the user credit analysis method provided in the embodiment of the present application Figure 4 .like Figure 4As shown, a possible implementation of the method includes the following steps:

[0116] S401, obtaining historical data of multiple influencing factors related to user credit analysis, and performing data cleaning and preprocessing.

[0117] Among them, historical data includes multiple influencing factors of user credit analysis and the corresponding specific values. Data cleaning and preprocessing include deleting missing values ​​or replacing them with estimated values, deleting or replacing outliers, and unifying the format of data types. Finally, duplicate data records are processed, only one key data record is retained, and other duplicate items are deleted.

[0118] S402, structurally processing the historical data of multiple influencing factors to obtain at least one two-dimensional format data.

[0119] In this step, the historical data of multiple influencing factors are structured, and an Excel table can be constructed based on the historical data, and the historical data can be filled into specified rows or columns to finally obtain two-dimensional format data with a unified format.

[0120] S403: performing association processing on at least one two-dimensional format data to obtain key factors.

[0121] In this step, the two-dimensional format data is subjected to association processing, which may be to analyze the two-dimensional format data using a preset tool to obtain influencing factors with strong correlation; the two-dimensional format data may also be subjected to statistics on the two-dimensional format data to determine the correlation coefficients between the influencing factors, obtain the correlation coefficients corresponding to the high correlation, and obtain the influencing factors corresponding to these correlation coefficients. Finally, the obtained influencing factors with strong correlation or the influencing factors corresponding to the high correlation are used as key factors.

[0122] S404, using the key factors as decision rows and decision columns of a cross decision table.

[0123] The decision rows of the cross decision table may be key factors of one part, and the decision columns of the cross decision table may be key factors of another part.

[0124] S405, configuring decision conditions for key factors and determining decision standard information.

[0125] Among them, decision conditions are configured for each key factor, for example, the decision conditions for the key factor of user age are configured as "greater than 60 years old", "greater than 30 years old and less than 60 years old" and "less than 30 years old". The decision standard information can be configured with weight coefficients for key factors, and finally obtained by weighted average according to different decision conditions corresponding to the key factors.

[0126] S406, constructing a cross decision table.

[0127] Among them, the construction of the cross decision table is based on key factors, decision conditions corresponding to the key factors and decision standard information.

[0128] S407, testing the cross decision table, and adjusting and optimizing the cross decision table according to the test result.

[0129] Among them, the validation data set is used to test the cross decision table, and the decision standard information in the cross decision table is compared with the score indicators corresponding to the historical data in the validation data set. The similarity between the compared data is compared, and the cross decision table is adjusted and optimized according to the test results, that is, the similarity of the data.

[0130] S408, monitor and evaluate the decision effect of the cross decision table and collect feedback data.

[0131] In this step, the decision criteria information evaluated by the cross decision table is set with two options of "accurate" and "inaccurate" for the user to confirm. The accuracy is calculated based on the "accurate" option selected by the user. If the accuracy is lower than 70%, the cross decision table needs to be re-optimized.

[0132] S409, regularly updating the cross decision table, and retraining and adjusting the cross decision table.

[0133] In this step, the cross decision table can be regularly updated according to feedback data or specific business scenarios to ensure that the decision process is always consistent with the actual situation. The cross decision table can be retrained and adjusted by regularly collecting full data related to specific businesses. The adjusted cross decision table is put into practical use.

[0134] S410, adding new key factors to the cross-decision table based on business needs, market trends or technological innovations.

[0135] In this step, if it is found that some users have a high decision-making standard information score, but their historical transaction records perform poorly, it means that the analysis process of the cross-decision table may be overfitting, and it is necessary to consider adding new key factors to optimize related business attributes. The association rule mining algorithm in machine learning can be used to discover the correlation and regularity hidden in the data, increase the correlation research between user credit information and historical transaction record performance, and judge the user's business capabilities and future unreliable factors.

[0136] S411, Use visualization tools and interactive techniques to display cross-decision tables.

[0137] In this step, visualization tools can be Data-Driven Documents (D3) and Baidu Echart. Visualization tools and interactive technologies are used to display cross-decision tables. By clicking on a user node, you can view its detailed income and expenditure records and related information with other users. At the same time, by displaying macro data such as the user's industry distribution and historical credit data, it can help assessment agencies conduct credit analysis and reduce the emergence of unreliable factors.

[0138] S412, introduce scenario simulation and prediction functions into the cross-decision table.

[0139] In this step, a specific business scenario is simulated, the user's credit is analyzed based on a cross-decision table, and future credit unreliability factors are predicted and evaluated based on historical data and current market conditions.

[0140] The user credit analysis method provided in the embodiment of the present application tests the constructed cross-decision table to obtain the test results, and then optimizes the cross-decision table according to the test results, finally obtaining the optimized cross-decision table, and monitoring and evaluating the cross-decision table, regularly updating it, and visually displaying it, and then introducing the scenario simulation and prediction functions into the cross-decision table. In this embodiment, the optimization and update of the cross-decision table improves the accuracy and reliability of the cross-decision table, and the introduction of the scenario simulation and prediction functions into the cross-decision table can realize the analysis of user credit in different scenarios, and can predict future credit unreliability factors.

[0141] Based on the above embodiments, the following are device embodiments involved in this application:

[0142] Figure 5 This is a schematic diagram of the structure of the user credit analysis device provided in the embodiment of the present application. Figure 5 As shown, the data processing device 500 includes:

[0143] An acquisition module 510 is used to acquire first data of a target influencing factor of an object to be analyzed;

[0144] A processing module 520 is used to input the first data of the target influencing factor into a pre-built cross decision table to obtain a credit score corresponding to the object to be analyzed, where the cross decision table is determined based on the historical data of multiple influencing factors and the correlation between the multiple influencing factors;

[0145] The determination module 530 is used to determine the credit analysis result of the object to be analyzed according to the credit score.

[0146] In an optional embodiment, before inputting the first data of the target influencing factor into the pre-built cross decision table to obtain the credit score corresponding to the object to be analyzed, the processing module 520 is further used to:

[0147] Structurally process the historical data of multiple influencing factors to obtain at least one two-dimensional format data;

[0148] Performing correlation processing on at least one two-dimensional format data to obtain key factors;

[0149] Configure decision conditions for key factors and determine decision criteria information;

[0150] A cross decision table is constructed based on key factors, decision conditions corresponding to key factors, and decision standard information.

[0151] In an optional embodiment, the processing module 520 performs association processing on at least one two-dimensional format data to obtain a key factor, which is specifically used for:

[0152] Using a preset tool, at least one two-dimensional format data is analyzed to obtain influencing factors with a correlation greater than a first preset threshold value, which are recorded as key factors.

[0153] In an optional embodiment, the processing module 520 performs association processing on at least one two-dimensional format data to obtain a key factor, which is specifically used for:

[0154] Performing statistics on at least one two-dimensional format data to determine the correlation coefficients between various influencing factors;

[0155] The influencing factors whose correlation coefficients are greater than the second preset threshold are recorded as key factors.

[0156] In an optional embodiment, the processing module 520 determines the decision criterion information, specifically for:

[0157] Configure weight coefficients for key factors;

[0158] According to the weight coefficient, the decision criterion information is determined.

[0159] In an optional embodiment, the processing module 520 is further configured to:

[0160] S1, test the cross decision table and obtain the test results;

[0161] S2, optimizing the cross decision table according to the test results to obtain an optimized cross decision table.

[0162] In an optional embodiment, the processing module 520 optimizes the cross decision table according to the test result to obtain an optimized cross decision table, which is specifically used for:

[0163] When the test result is lower than the first threshold, the cross decision table is adjusted, and the above steps S1-S2 are repeated until the test result is greater than or equal to the first threshold. The obtained cross decision table is recorded as the optimized cross decision table.

[0164] Based on the above embodiments, Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 6 As shown, the electronic device 600 includes: a processor 610, a memory 620 and a bus 630;

[0165] The memory 620 is used to store computer-executable instructions of the processor 610;

[0166] The processor 610 is configured to execute the technical solution of any of the aforementioned method embodiments by executing computer execution instructions.

[0167] Optionally, the memory 620 may be independent or integrated with the processor 610 .

[0168] Optionally, the memory 620 may include a random access memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0169] The bus 630 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the drawings of the present application, but it does not mean that there is only one bus or one type of bus.

[0170] The above-mentioned processor can be a general-purpose processor, including a central processing unit CPU, a network processor (NP), etc.; it can also be a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0171] The electronic device is used to execute the technical solution of any of the aforementioned method embodiments, and its implementation principle and technical effect are similar and will not be repeated here.

[0172] An embodiment of the present application also provides a computer-readable storage medium on which computer execution instructions are stored. When the computer execution instructions are executed by a processor, they are used to implement the technical solution provided by any of the above method embodiments.

[0173] An embodiment of the present application also provides a computer program product, including a computer program, which includes computer instructions stored in a computer-readable storage medium. When the computer program is executed by a processor, it is used to implement the technical solution provided by any of the above method embodiments.

[0174] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0175] It should be further noted that, although the various steps in the flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flow chart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0176] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0177] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0178] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory (RRAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), an enhanced dynamic random access memory (EDRAM), a high-bandwidth memory (HBM), a hybrid memory cube (HMC), etc.

[0179] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0180] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0181] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0182] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A user credit analysis method, characterized in that: include: Acquire first data of a target influencing factor of the object to be analyzed; Inputting the first data of the target influencing factor into a pre-constructed cross decision table to obtain a credit score corresponding to the object to be analyzed, wherein the cross decision table is determined based on historical data of multiple influencing factors and correlations between the multiple influencing factors; A credit analysis result of the object to be analyzed is determined according to the credit score.

2. The method according to claim 1, characterized in that Before inputting the first data of the target influencing factor into a pre-constructed cross decision table to obtain the credit score corresponding to the object to be analyzed, the method further includes: Structurally processing the historical data of the multiple influencing factors to obtain at least one two-dimensional format data; Performing association processing on the at least one two-dimensional format data to obtain a key factor; configuring decision conditions for the key factors and determining decision standard information; The cross decision table is constructed according to the key factors, the decision conditions corresponding to the key factors, and the decision standard information.

3. The method according to claim 2, characterized in that The step of performing association processing on the at least one two-dimensional format data to obtain the key factor includes: The at least one two-dimensional format data is analyzed by using a preset tool to obtain influencing factors with a correlation greater than a first preset threshold value, which are recorded as the key factors.

4. The method according to claim 2, characterized in that: The step of performing association processing on the at least one two-dimensional format data to obtain the key factor includes: Performing statistics on the at least one two-dimensional format data to determine the correlation coefficient between various influencing factors; The influencing factors whose correlation coefficients are greater than the second preset threshold are recorded as the key factors.

5. The method according to any one of claims 2 to 4, characterized in that: The information for determining the decision criteria includes: configuring weight coefficients for the key factors; The decision criterion information is determined according to the weight coefficient.

6. The method according to claim 1 or 2, characterized in that: The method further comprises: S1, testing the cross decision table to obtain a test result; S2, optimizing the cross decision table according to the test result to obtain an optimized cross decision table.

7. The method according to claim 6, characterized in that The step of optimizing the cross decision table according to the test result to obtain an optimized cross decision table includes: When the test result is lower than the first threshold, the cross decision table is adjusted, and the above steps S1-S2 are repeated until the test result is greater than or equal to the first threshold. The obtained cross decision table is recorded as an optimized cross decision table.

8. A user credit analysis device, characterized in that: include: An acquisition module, used to acquire first data of a target influencing factor of the object to be analyzed; A processing module, configured to input the first data of the target influencing factor into a pre-constructed cross-decision table to obtain a credit score corresponding to the object to be analyzed, wherein the cross-decision table is determined based on historical data of multiple influencing factors and correlations between the multiple influencing factors; A determination module is used to determine the credit analysis result of the object to be analyzed according to the credit score.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1 to 7.

Images