Dynamic credit assessment method and device, electronic equipment and storage medium

Abstract

The invention provides a dynamic credit assessment method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the behavior data of a user in corresponding multiple dimensions from a multi-dimensional data source end which affects the credit of the user in a recent preset time period in real time, carrying out the exception processing of the behavior data, and obtaining the behavior data of the user in corresponding multiple dimensions; the method comprises the steps of obtaining target behavior data suitable for credit evaluation, determining a multi-factor analysis result influencing user credit by performing behavior analysis on the target behavior data, and dynamically generating a credit score of a user at the current time based on the target behavior data, the multi-factor analysis result influencing the user credit and a preset credit scoring model. According to the dynamic credit assessment method, the assessment period is shortened, the credit score can reflect the current credit condition of the user in time, the hysteresis problem of the credit score of the user is solved, the bank is prevented from missing high-quality customers or excessive lending to the user with high credit risk is avoided, the credit risk is reduced, and the credit efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a dynamic credit assessment method, device, electronic device and storage medium. Background Art

[0002] With the deepening of global economic integration and the rapid development of financial markets, the accuracy and efficiency of credit assessment are crucial for financial institutions, enterprises and even individuals. The increasing volatility of financial markets, the continuous innovation of financial products and the complexity of the global economic environment have made credit risk one of the main risks faced by financial institutions. In order to effectively control risks and ensure the safety of funds, it is necessary to conduct credit assessment on users.

[0003] Currently, users' credit is mainly assessed through historical credit records and financial statements.

[0004] However, the existing credit assessment methods have a long assessment cycle, and credit scores are difficult to reflect the user's current credit status in a timely manner. They have a certain lag, which may cause banks to miss out on high-quality customers or over-lending to users with high credit risks, thereby increasing credit risks and reducing credit efficiency. Summary of the invention

[0005] In view of this, the purpose of the present application is to provide a dynamic credit assessment method, device, electronic device and storage medium, which can determine the multi-factor analysis results that affect the user's credit by performing real-time behavioral analysis on the user's behavioral data in multiple dimensions, and dynamically generate the user's current credit score based on the target behavioral data, the multi-factor analysis results and the credit scoring model, thereby shortening the assessment cycle, and the credit score can timely reflect the user's current credit status, solving the lag problem of the user's credit score, avoiding banks missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0006] In a first aspect, an embodiment of the present application provides a dynamic credit assessment method, the method comprising:

[0007] Real-time acquisition of the user's behavior data in corresponding multiple dimensions from the multi-dimensional data source that affects the user's credit in the most recent preset time period;

[0008] Performing abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment;

[0009] By performing behavioral analysis on the target behavior data, a multi-factor analysis result affecting the user's credit is determined; wherein the multi-factor analysis result includes analysis results corresponding to behavioral factors and financial factors respectively;

[0010] Based on the target behavior data, the analysis results of the multiple factors affecting the user's credit, and a preset credit scoring model, the user's current credit score is dynamically generated.

[0011] In a possible implementation manner, performing exception processing on the behavior data to obtain target behavior data suitable for credit assessment includes:

[0012] Performing anomaly detection on the behavior data to determine various types of abnormal data in the behavior data; the various types of abnormal data at least include data corresponding to a duplication type, a missing type, and an error type, respectively;

[0013] According to the abnormal information of each type of abnormal data, a cleaning method for each type of abnormal data is determined, and the abnormal data of each type is cleaned using the cleaning method corresponding to the type to obtain target behavior data suitable for credit assessment.

[0014] In a possible implementation, the repetition type includes a transaction repetition type; and the determining each type of abnormal data in the behavior data includes:

[0015] Obtaining a preset transaction duplication screening threshold; wherein the transaction duplication screening threshold at least includes a transaction time threshold;

[0016] In response to the existence of at least two transaction records in the behavior data, the difference in transaction time between which is within the transaction time threshold and the transaction amount is the same, corresponding transaction duplication data is determined based on the transaction record.

[0017] In a possible implementation, the method further includes:

[0018] Generate corresponding credit decision suggestions for preset target financial businesses based on the credit score and preset business rules; wherein the target financial businesses at least include loan approval and credit limit adjustment;

[0019] The credit decision suggestion is integrated into a corresponding credit management system so that corresponding processing is performed in the credit management system based on the credit decision suggestion.

[0020] In a possible implementation, the determining of the multi-factor analysis result affecting the user's credit by performing behavior analysis on the target behavior data includes:

[0021] Pre-configure a big data processing model based on the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data; wherein the big data processing model is used for stream processing;

[0022] Based on the abnormal indicators corresponding to each factor affecting user credit and the configured big data processing model, the target behavior data is monitored under each abnormal indicator to obtain the corresponding multi-factor analysis results.

[0023] In a possible implementation, the credit scoring model is obtained by the following steps:

[0024] Obtaining a machine learning model and historical behavior data of the user within a preset historical time period; wherein the historical behavior data is historical behavior data after streaming processing;

[0025] The machine learning model is trained based on the historical behavior data to obtain the credit scoring model.

[0026] In a possible implementation, the method further includes:

[0027] In response to an appeal operation filed by the user when determining that the credit score is incorrect, updating the credit score in real time based on the appeal data submitted by the user;

[0028] The credit scoring model is calibrated in real time based on the complaint data to update the credit scoring model in real time.

[0029] In a second aspect, an embodiment of the present application further provides a dynamic credit assessment device, the device comprising:

[0030] The first acquisition module is used to obtain the user's behavior data in corresponding multiple dimensions in real time from the multi-dimensional data source that affects the user's credit in the latest preset time period;

[0031] A second acquisition module is used to perform abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment;

[0032] A determination module, configured to determine a multi-factor analysis result affecting the user's credit by performing a behavior analysis on the target behavior data; wherein the multi-factor analysis result includes analysis results corresponding to the behavior factor and the financial factor respectively;

[0033] The scoring module is used to dynamically generate the user's current credit score based on the target behavior data, the multi-factor analysis results that affect the user's credit, and a preset credit scoring model.

[0034] In a possible implementation manner, the second acquisition module is specifically configured to:

[0035] Performing anomaly detection on the behavior data to determine various types of abnormal data in the behavior data; the various types of abnormal data at least include data corresponding to a duplication type, a missing type, and an error type, respectively;

[0036] According to the abnormal information of each type of abnormal data, a cleaning method for each type of abnormal data is determined, and the abnormal data of each type is cleaned using the cleaning method corresponding to the type to obtain target behavior data suitable for credit assessment.

[0037] In a possible implementation, the repetition type includes a transaction repetition type; and the determination module is specifically configured to:

[0038] Obtaining a preset transaction duplication screening threshold; wherein the transaction duplication screening threshold at least includes a transaction time threshold;

[0039] In response to the existence of at least two transaction records in the behavior data, the difference in transaction time between which is within the transaction time threshold and the transaction amount is the same, corresponding transaction duplication data is determined based on the transaction record.

[0040] In a possible implementation manner, the dynamic credit assessment device further includes:

[0041] A decision module, used to generate corresponding credit decision suggestions for preset target financial services based on the credit score and preset business rules; wherein the target financial services at least include loan approval and credit limit adjustment;

[0042] An integration module is used to integrate the credit decision suggestion into a corresponding credit management system so as to perform corresponding processing based on the credit decision suggestion in the credit management system.

[0043] In a possible implementation manner, the determining module is specifically configured to:

[0044] Pre-configure a big data processing model based on the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data; wherein the big data processing model is used for stream processing;

[0045] Based on the abnormal indicators corresponding to each factor affecting user credit and the configured big data processing model, the target behavior data is monitored under each abnormal indicator to obtain the corresponding multi-factor analysis results.

[0046] In a possible implementation, the credit scoring model is obtained by the following steps:

[0047] Obtaining a machine learning model and historical behavior data of the user within a preset historical time period; wherein the historical behavior data is historical behavior data after streaming processing;

[0048] The machine learning model is trained based on the historical behavior data to obtain the credit scoring model.

[0049] In a possible implementation manner, the dynamic credit assessment device further includes:

[0050] A first updating module, configured to respond to a complaint operation filed by the user when the credit score is judged to be incorrect, and update the credit score in real time based on the complaint data submitted by the user;

[0051] The second updating module is used to calibrate the credit scoring model in real time based on the complaint data to update the credit scoring model in real time.

[0052] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the dynamic credit assessment method as described in any one of the first aspects.

[0053] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the dynamic credit assessment method described in any one of the first aspects are executed.

[0054] The embodiment of the present application provides a dynamic credit assessment method, device, electronic device and storage medium, which respectively obtains the user's behavior data in the corresponding multiple dimensions from the multi-dimensional data source end that affects the user's credit in the recent preset time period, performs abnormal processing on the behavior data, obtains the target behavior data suitable for credit assessment, and determines the multi-factor analysis results that affect the user's credit by performing behavioral analysis on the target behavior data. Based on the target behavior data, the multi-factor analysis results that affect the user's credit and the pre-set credit scoring model, the user's credit score at the current time is dynamically generated. The dynamic credit assessment method of the present application determines the multi-factor analysis results that affect the user's credit by performing real-time behavioral analysis on the user's behavior data in multiple dimensions, and dynamically generates the user's credit score at the current time based on the target behavior data, the multi-factor analysis results and the credit scoring model, shortening the assessment cycle, and the credit score can timely reflect the user's current credit status, solving the lag problem of the user's credit score, avoiding the bank from missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0055] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 is a flow chart of a dynamic credit assessment method according to an embodiment of the present application;

[0058] Figure 2 is a schematic diagram of the dynamic credit assessment process;

[0059] Figure 3 is a flow chart of a dynamic credit assessment method according to another embodiment of the present application;

[0060] Figure 4 is a flow chart of a dynamic credit assessment method according to another embodiment of the present application;

[0061] Figure 5 is a flow chart of a dynamic credit assessment method according to another embodiment of the present application;

[0062] Figure 6 is a flow chart of a dynamic credit assessment method according to another embodiment of the present application;

[0063] Figure 7 is a structural schematic diagram of a dynamic credit evaluation device according to an embodiment of the present application;

[0064] Figure 8 It is a structural diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0065] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0066] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0067] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0068] With the deepening of global economic integration and the rapid development of financial markets, the accuracy and efficiency of credit assessment are crucial for financial institutions, enterprises and even individuals. The increasing volatility of financial markets, the continuous innovation of financial products and the complexity of the global economic environment have made credit risk one of the main risks faced by financial institutions. In order to effectively control risks and ensure the safety of funds, it is necessary to conduct credit assessment on users.

[0069] At present, credit assessment of users is mainly carried out through historical credit records and financial statements. However, the existing credit assessment method has a long assessment cycle, and the credit score is difficult to reflect the user's current credit status in a timely manner. It has a certain lag, which may cause banks to miss out on high-quality customers or over-lend to users with high credit risks, thereby increasing credit risks and reducing credit efficiency.

[0070] To address this problem, the present application provides a dynamic credit assessment method, device, electronic device and storage medium, which determines the multi-factor analysis results that affect the user's credit by performing real-time behavioral analysis on the user's behavioral data in multiple dimensions, and dynamically generates the user's current credit score based on the target behavioral data, the multi-factor analysis results and the credit scoring model, thereby shortening the assessment cycle, and the credit score can promptly reflect the user's current credit status, solving the lag problem of the user's credit score, avoiding banks missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0071] Figure 1 is a flow chart of a dynamic credit assessment method according to an embodiment of the present application. Figure 1 As shown, the dynamic credit assessment method of the embodiment of the present application may specifically include:

[0072] S101, obtaining the user's behavior data in corresponding multiple dimensions in real time from the multi-dimensional data source that affects the user's credit in the most recent preset time period.

[0073] S102, performing abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment.

[0074] S103, determining the results of a multi-factor analysis that affects the user's credit by performing a behavior analysis on the target behavior data.

[0075] S104, dynamically generating the user's current credit score based on the target behavior data, the analysis results of multiple factors affecting the user's credit, and a pre-set credit score model.

[0076] The above-mentioned dynamic credit assessment method determines the multi-factor analysis results that affect the user's credit by conducting real-time behavioral analysis on the user's behavioral data in multiple dimensions, and dynamically generates the user's current credit score based on the target behavioral data, multi-factor analysis results and credit scoring model, thereby shortening the assessment cycle. The credit score can timely reflect the user's current credit status, solving the lag problem of the user's credit score, and avoiding banks missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0077] The above exemplary steps of the embodiment of the present application are described below with reference to specific examples:

[0078] S101, obtaining the user's behavior data in corresponding multiple dimensions in real time from the multi-dimensional data source that affects the user's credit in the most recent preset time period.

[0079] In the embodiment of the present application, the user is the user to be evaluated for credit, the multi-dimensional data source is the collection end of the multi-dimensional data source that affects the user's credit, and the behavior data is the data required for the user's credit evaluation. The behavior data includes at least financial records, transaction history, public records, social media activities, and geographic location data. The user's behavior data in the corresponding multi-dimensional areas is obtained from the multi-dimensional data source that affects the user's credit in real time within the recent preset time period for subsequent processing. For example, Figure 2 As shown, behavioral data is collected through a data collection module.

[0080] Optionally, an automated data crawling tool may be used to obtain user behavior data in corresponding multiple dimensions from a multi-dimensional data source in real time within a recent preset time period.

[0081] S102, performing abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment.

[0082] In the embodiment of the present application, the target behavior data is the target behavior data suitable for credit evaluation of the user, and the behavior data of the user in the corresponding multi-dimensional acquired in step S101 is processed abnormally to obtain the target behavior data for evaluation for subsequent processing. Figure 2 As shown, after preprocessing the behavior data, that is, exception processing, the behavior data can be processed into target behavior data.

[0083] S103, determining the results of a multi-factor analysis that affects the user's credit by performing a behavior analysis on the target behavior data.

[0084] In the embodiment of the present application, the multi-factor analysis results include analysis results corresponding to behavioral factors and financial factors. For example, the multi-factors may include behavioral factors and financial factors such as overdue payments, income changes, and abnormal consumption behaviors. By performing behavioral analysis on the target behavior data obtained in step S102, the multi-factor analysis results affecting the user's credit are determined for subsequent processing. Figure 2 As shown, the real-time analysis engine performs behavioral analysis on the target behavior data to determine the multi-factor analysis results that affect the user's credit.

[0085] S104, dynamically generating the user's current credit score based on the target behavior data, the analysis results of multiple factors affecting the user's credit, and a pre-set credit score model.

[0086] In the embodiment of the present application, the credit scoring model is a pre-set credit scoring model for evaluating the credit of the user. Based on the target behavior data obtained in step S102 and the multi-factor analysis results affecting the user's credit obtained in step S103 and the credit scoring model, the credit score of the user at the current time is dynamically generated to complete the real-time credit evaluation of the user. For example, Figure 2 As shown, the user's current credit is dynamically scored according to the credit scoring model.

[0087] The dynamic credit assessment method provided by the embodiment of the present application obtains the user's behavior data in the corresponding multiple dimensions from the multi-dimensional data source end that affects the user's credit in the recent preset time period in real time, performs abnormal processing on the behavior data, obtains the target behavior data suitable for credit assessment, and determines the multi-factor analysis results that affect the user's credit by performing behavioral analysis on the target behavior data. Based on the target behavior data, the multi-factor analysis results that affect the user's credit, and the pre-set credit scoring model, the user's current credit score is dynamically generated. The dynamic credit assessment method of the present application determines the multi-factor analysis results that affect the user's credit by performing real-time behavioral analysis on the user's behavior data in multiple dimensions, and dynamically generates the user's current credit score based on the target behavior data, the multi-factor analysis results, and the credit scoring model, shortening the assessment cycle, and the credit score can timely reflect the user's current credit status, solving the lag problem of the user's credit score, avoiding the bank from missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0088] Further, such as Figure 3 As shown, step S102 in the above embodiment, “performing abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment”, may specifically include the following steps:

[0089] S301, performing anomaly detection on the behavior data to determine various types of abnormal data in the behavior data.

[0090] In the embodiment of the present application, each type of abnormal data includes at least data corresponding to a duplication type, a missing type, and an error type, respectively. The behavior data in the above embodiment is subjected to abnormality detection to determine each type of abnormal data in the behavior data for subsequent processing. Figure 2 As shown, during data preprocessing, data cleaning is performed, and each type of abnormal data is duplicates, missing values, and erroneous data. It should be noted that the duplicate type is the data type with duplicates, the missing type is the data type with missing values, and the erroneous type is the data type with erroneous information.

[0091] S302, determining a cleaning method for each type of abnormal data according to abnormal information of each type of abnormal data, and using the cleaning method corresponding to each type to clean the abnormal data of the abnormal type to obtain target behavior data suitable for credit assessment.

[0092] In the embodiment of the present application, according to the abnormal information of each type of abnormal data obtained in step S301, the cleaning method of each type of abnormal data is determined, and the abnormal data of the abnormal type is cleaned using the cleaning method corresponding to each type to obtain target behavior data suitable for credit assessment. For example, according to the abnormal information of the repeated type abnormal data, that is, the repeated items, it is determined that the cleaning method of the repeated type abnormal data is to remove the repeated items of the repeated type abnormal data, and the same is true for other types of abnormal data, and the missing types are filled and the wrong types are corrected, so as to obtain target behavior data suitable for credit assessment. For example, Figure 2 As shown, during data cleaning, duplicates are removed, missing values ​​are filled, and erroneous data are corrected.

[0093] Therefore, by performing exception processing on the behavior data, the cleaning of the behavior data is achieved.

[0094] Furthermore, the target behavior data is formatted to convert the target behavior data into target behavior data in a unified format.

[0095] In the embodiment of the present application, the target behavior data obtained after the exception processing is formatted to convert the target behavior data into target behavior data in a unified format for easy subsequent processing. Figure 2 As shown, during data preprocessing, data formatting is performed, that is, the data is converted into a unified format.

[0096] Therefore, the preprocessing of the behavioral data is achieved by cleaning and formatting the behavioral data.

[0097] Further, the repetition type includes transaction repetition types such as Figure 4 As shown, the “determining various types of abnormal data in the behavior data” in step S301 in the above embodiment may specifically include the following steps:

[0098] S401, obtaining a preset transaction duplication screening threshold.

[0099] In an embodiment of the present application, the transaction duplication screening threshold includes at least a transaction time threshold. For the transaction duplication type data in the duplication type abnormal data, a preset transaction duplication screening threshold for the transaction duplication type abnormal data is obtained for subsequent processing.

[0100] S402: In response to the existence of at least two transaction records in the behavior data, the difference in transaction time between which is within the transaction time threshold and the transaction amounts are the same, determining corresponding transaction duplication data based on the transaction records.

[0101] In the embodiment of the present application, the behavior data in the above embodiment is judged, and when there are at least two transaction records in the behavior data whose transaction time difference is within the transaction time threshold and the transaction amount is the same, that is, the transaction amount of these transaction records is the same and the transaction time is close (within the transaction time threshold), then the corresponding abnormal data of the transaction duplication type is determined based on the transaction record. For example, if the transaction time threshold is 2S, there are two transaction records with the same transaction amount and the difference in transaction time between the two transaction records is not greater than 2S, then it is determined as abnormal data of the transaction duplication type.

[0102] Furthermore, a corresponding credit decision suggestion is generated for a preset target financial business based on the credit score and preset business rules; the credit decision suggestion is integrated into the corresponding credit management system so that corresponding processing is performed in the credit management system based on the credit decision suggestion. The target financial business includes at least loan approval and credit limit adjustment. For example, Figure 2 As shown, in the decision support module, credit scores and other business rules are combined to provide decision-making basis, namely credit decision suggestions, for target financial businesses such as loan approval and credit limit adjustment. The decision suggestions are integrated into the bank's credit management system and fed back to the bank system.

[0103] Further, such as Figure 5 As shown, the step S103 in the above embodiment of "determining the multi-factor analysis result affecting the user's credit by analyzing the target behavior data" may specifically include the following steps:

[0104] S501, pre-configure a big data processing model based on the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data.

[0105] In the embodiment of the present application, the abnormal monitoring rules are rules for monitoring abnormal behavior data set for behavior data, and the big data processing model is used for streaming processing (Flink, streaming data processing). The big data processing model is pre-configured according to the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data for subsequent processing. Among them, the big data processing model can be a big data processing framework, such as Apache Kafka, Apache Flink, etc., and the big data processing model is configured according to the user's behavior data in multiple dimensions and the abnormal monitoring rules in the above embodiments. For example, Figure 2 As shown, in the real-time analysis engine, streaming data processing is performed on the behavior data.

[0106] S502, based on the abnormal indicators corresponding to each factor affecting the user's credit and the configured big data processing model, the target behavior data is monitored under each abnormal indicator to obtain corresponding multi-factor analysis results.

[0107] In the embodiment of the present application, the abnormal index is the abnormal index corresponding to each factor affecting the user's credit. Based on the abnormal index corresponding to each factor affecting the user's credit and the big data processing model, the target behavior data is monitored under each abnormal index to obtain the corresponding multi-factor analysis results. Figure 2 As shown, the target behavior data is monitored in real time, and factors that may affect the user's credit score, such as overdue payments, income changes, abnormal consumption behavior, etc., are continuously monitored.

[0108] Therefore, the user's credit score is generated instantly through real-time analysis of the target behavior data, namely streaming processing and real-time monitoring.

[0109] Furthermore, the credit scoring model is obtained through the following steps:

[0110] Obtain a machine learning model and the user's historical behavior data within a preset historical time period; train the machine learning model based on the historical behavior data to obtain a credit scoring model. The historical behavior data is the historical behavior data after streaming processing, that is, the historical behavior data processed by the configured big data processing model as described in the above embodiment. In short, use the user's historical data to train a machine learning model, such as a neural network, to obtain a credit scoring model to predict credit risk. For example, the preset historical time period is one month, and the neural network model and the user's historical behavior data within one month are obtained for training to obtain a credit scoring model.

[0111] Further, such as Figure 6 As shown, the dynamic credit assessment method of the embodiment of the present application may also include the following steps:

[0112] S601, in response to a complaint operation filed by a user when the credit score is judged to be incorrect, the credit score is updated in real time based on the complaint data submitted by the user.

[0113] In an embodiment of the present application, when the user determines that the credit score is incorrect, an appeal operation is filed, and the credit score is updated in real time based on the appeal data submitted by the user in the appeal operation to obtain an updated credit score.

[0114] S602, calibrating the credit scoring model in real time based on the complaint data to update the credit scoring model in real time.

[0115] In the embodiment of the present application, the credit scoring model is calibrated in real time based on the complaint data in step S601 to update the credit scoring model in real time. Optionally, the credit scoring model may be calibrated and updated regularly.

[0116] Further, in response to the user's viewing operation on the target interface in the credit management system, the credit score is displayed on the target interface for the user to view. In short, the user is allowed to view his or her own credit score.

[0117] Therefore, credit scores and credit scoring models are updated in real time through user complaints, which represents the user feedback and self-correction mechanism and improves the accuracy of credit scores and credit scoring models.

[0118] Figure 7 is a flow chart of a dynamic credit evaluation device according to an embodiment of the present application. Figure 7 As shown, the dynamic credit evaluation device 700 of the embodiment of the present application may specifically include:

[0119] The first acquisition module 701 is used to acquire the user's behavior data in corresponding multiple dimensions in real time from the multi-dimensional data source that affects the user's credit in the latest preset time period.

[0120] The second acquisition module 702 is used to perform abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment.

[0121] The determination module 703 is used to determine the multi-factor analysis results affecting the user's credit by performing behavior analysis on the target behavior data; wherein the multi-factor analysis results include analysis results corresponding to the behavior factors and the financial factors respectively.

[0122] The scoring module 704 is used to dynamically generate the user's current credit score based on the target behavior data, the analysis results of multiple factors affecting the user's credit, and a preset credit scoring model.

[0123] In a possible implementation manner, the second acquisition module is specifically configured to:

[0124] Performing anomaly detection on the behavior data to determine various types of abnormal data in the behavior data; each type of abnormal data at least includes data corresponding to a duplication type, a missing type, and an error type;

[0125] According to the abnormal information of each type of abnormal data, a cleaning method for each type of abnormal data is determined, and the abnormal data of each type is cleaned using the cleaning method corresponding to the type to obtain target behavior data suitable for credit assessment.

[0126] In a possible implementation, the repetition type includes a transaction repetition type; and the determination module is specifically configured to:

[0127] Obtaining a preset transaction duplication screening threshold; wherein the transaction duplication screening threshold at least includes a transaction time threshold;

[0128] In response to the existence of at least two transaction records in the behavior data, the difference in transaction time between which is within a transaction time threshold and the transaction amounts are the same, corresponding transaction duplication data is determined based on the transaction records.

[0129] In a possible implementation manner, the dynamic credit assessment device further includes:

[0130] A decision module, which is used to generate corresponding credit decision suggestions for preset target financial businesses based on credit scores and preset business rules; wherein the target financial businesses at least include loan approval and credit limit adjustment;

[0131] The integration module is used to integrate the credit decision suggestion into the corresponding credit management system so as to perform corresponding processing based on the credit decision suggestion in the credit management system.

[0132] In a possible implementation, the determination module is specifically configured to:

[0133] Pre-configure a big data processing model based on the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data; wherein the big data processing model is used for stream processing;

[0134] Based on the abnormal indicators corresponding to each factor affecting user credit and the configured big data processing model, the target behavior data is monitored under each abnormal indicator to obtain the corresponding multi-factor analysis results.

[0135] In a possible implementation, the credit scoring model is obtained by the following steps:

[0136] Obtain the machine learning model and the user's historical behavior data within a preset historical time period; wherein the historical behavior data is the historical behavior data after streaming processing;

[0137] The machine learning model is trained based on historical behavioral data to obtain a credit scoring model.

[0138] In a possible implementation manner, the dynamic credit assessment device further includes:

[0139] A first updating module, for responding to a complaint operation filed by a user when the credit score is judged to be incorrect, and updating the credit score in real time based on the complaint data submitted by the user;

[0140] The second updating module is used to calibrate the credit scoring model in real time based on the complaint data to update the credit scoring model in real time.

[0141] The dynamic credit evaluation device provided in the embodiment of the present application obtains the user's behavior data in the corresponding multiple dimensions from the multi-dimensional data source end that affects the user's credit in the recent preset time period in real time, performs abnormal processing on the behavior data, obtains the target behavior data suitable for credit evaluation, and determines the multi-factor analysis results that affect the user's credit by performing behavior analysis on the target behavior data. Based on the target behavior data, the multi-factor analysis results that affect the user's credit, and the pre-set credit scoring model, the user's current credit score is dynamically generated. The dynamic credit evaluation device of the present application determines the multi-factor analysis results that affect the user's credit by performing real-time behavior analysis on the user's behavior data in multiple dimensions, and dynamically generates the user's current credit score based on the target behavior data, the multi-factor analysis results, and the credit scoring model, shortening the evaluation cycle, and the credit score can timely reflect the user's current credit status, solving the lag problem of the user's credit score, avoiding the bank from missing out on high-quality customers or over-lending to users with high credit risks, thereby reducing credit risks and improving credit efficiency.

[0142] like Figure 8 As shown, an electronic device 800 provided in an embodiment of the present application includes: a processor 801, a memory 802 and a bus, wherein the memory 802 stores machine-readable instructions executable by the processor 801. When the electronic device is running, the processor 801 communicates with the memory 802 through the bus, and the processor 801 executes the machine-readable instructions to perform the steps of the dynamic credit assessment method as described above.

[0143] Specifically, the memory 802 and the processor 801 can be general-purpose memories and processors, which are not specifically limited here. When the processor 801 runs the computer program stored in the memory 802, the dynamic credit assessment method can be executed.

[0144] Corresponding to the above-mentioned dynamic credit assessment method, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned dynamic credit assessment method are executed.

[0145] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0146] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0147] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0148] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the deployment method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0149] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A dynamic credit assessment method, characterized in that: The method comprises: Real-time acquisition of the user's behavior data in corresponding multiple dimensions from the multi-dimensional data source that affects the user's credit in the most recent preset time period; Performing abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment; By performing behavioral analysis on the target behavior data, a multi-factor analysis result affecting the user's credit is determined; wherein the multi-factor analysis result includes analysis results corresponding to behavioral factors and financial factors respectively; Based on the target behavior data, the analysis results of the multiple factors affecting the user's credit, and a preset credit scoring model, the user's current credit score is dynamically generated.

2. The method according to claim 1, characterized in that The performing of abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment includes: Performing anomaly detection on the behavior data to determine various types of abnormal data in the behavior data; the various types of abnormal data at least include data corresponding to a duplication type, a missing type, and an error type, respectively; According to the abnormal information of each type of abnormal data, a cleaning method for each type of abnormal data is determined, and the abnormal data of the abnormal type is cleaned using the cleaning method corresponding to each type to obtain target behavior data suitable for credit assessment.

3. The method according to claim 2, characterized in that The repetition type includes a transaction repetition type; and the determining of each type of abnormal data in the behavior data includes: Obtaining a preset transaction duplication screening threshold; wherein the transaction duplication screening threshold at least includes a transaction time threshold; In response to the existence of at least two transaction records in the behavior data, the difference in transaction time between which is within the transaction time threshold and the transaction amount is the same, corresponding transaction duplication data is determined based on the transaction record.

4. The method according to claim 1, characterized in that The method further comprises: Generate corresponding credit decision suggestions for preset target financial businesses based on the credit score and preset business rules; wherein the target financial businesses at least include loan approval and credit limit adjustment; The credit decision suggestion is integrated into a corresponding credit management system so that corresponding processing is performed in the credit management system based on the credit decision suggestion.

5. The method according to claim 1, characterized in that: The method of performing behavioral analysis on the target behavior data to determine the results of a multi-factor analysis affecting the user's credit includes: Pre-configure a big data processing model based on the characteristics of the user's behavior data in multiple dimensions and the abnormal monitoring rules of the behavior data; wherein the big data processing model is used for stream processing; Based on the abnormal indicators corresponding to each factor affecting user credit and the configured big data processing model, the target behavior data is monitored under each abnormal indicator to obtain the corresponding multi-factor analysis results.

6. The method according to claim 1, characterized in that The credit scoring model is obtained through the following steps: Obtaining a machine learning model and historical behavior data of the user within a preset historical time period; wherein the historical behavior data is historical behavior data after streaming processing; The machine learning model is trained based on the historical behavior data to obtain the credit scoring model.

7. The method according to claim 4, characterized in that The method further comprises: In response to an appeal operation filed by the user when determining that the credit score is incorrect, updating the credit score in real time based on the appeal data submitted by the user; The credit scoring model is calibrated in real time based on the complaint data to update the credit scoring model in real time.

8. A dynamic credit assessment device, characterized in that: The device comprises: The first acquisition module is used to obtain the user's behavior data in corresponding multiple dimensions in real time from the multi-dimensional data source that affects the user's credit in the latest preset time period; A second acquisition module is used to perform abnormal processing on the behavior data to obtain target behavior data suitable for credit assessment; A determination module, configured to determine a multi-factor analysis result affecting the user's credit by performing a behavior analysis on the target behavior data; wherein the multi-factor analysis result includes analysis results corresponding to the behavior factor and the financial factor respectively; The scoring module is used to dynamically generate the user's current credit score based on the target behavior data, the multi-factor analysis results that affect the user's credit, and a preset credit scoring model.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the dynamic credit assessment method as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the dynamic credit assessment method as claimed in any one of claims 1 to 7.

Images