Method, device and storage medium for determining credit score of loan users

Abstract

This application discloses a method, apparatus, and storage medium for determining a loan user's credit score, relating to the field of financial technology. The method includes: acquiring N types of financial data of the loan user, with the user's authorization, where N is an integer greater than 1; calculating mathematical statistics for the i-th type of financial data if missing values ​​and / or outliers are detected in the i-th type of financial data; correcting the missing values ​​and / or outliers in the i-th type of financial data based on the mathematical statistics, and extracting features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; and determining the loan user's target credit score based on the N sets of feature sequences corresponding to the N types of financial data. This application solves the technical problem in the prior art where the accuracy of determining a user's credit score is low due to the large number of types of user data.

Description

Technical Field

[0001] This application relates to the field of financial technology, and more specifically, to a method, apparatus, and storage medium for determining the credit score of a loan user. Background Technology

[0002] With the continuous development of the economy and society, while the financial industry's products and services are improving efficiency and serving society, they also face many challenges, the most prominent of which is how to effectively control risks. In bank lending, personal loan credit assessment is one of the important measures for risk control. Personal loan credit assessment can effectively identify customer default risk, prevent credit risk in advance, and reduce credit risk.

[0003] Currently, banks primarily rely on manual review for personal loan credit assessments. This involves personnel manually reviewing loan application materials submitted by customers. While manual review is less costly, it cannot cope with the significant increase in personal loan demand and leads to inefficiency in bank lending operations. Furthermore, inconsistent review standards and subjective judgment biases cannot guarantee the fairness and accuracy of the review results. Therefore, banks have adopted automated personal loan credit assessment systems that supplement manual review.

[0004] Currently, banks primarily employ rule-based personal loan credit assessment models. These models utilize rule bases and rule engines to review loan applications submitted by customers based on the rules within the rule base. While this approach is relatively low-cost, it often struggles to effectively identify customer default risks when processing large-scale, diverse user data, leading to technical issues such as low accuracy in determining user credit scores.

[0005] There is currently no effective solution to the above problems. Summary of the Invention

[0006] This application provides a method, apparatus, and storage medium for determining the credit score of a loan user, in order to at least solve the technical problem in the prior art where the accuracy of determining a user's credit score is low due to the large number of types of user data.

[0007] According to one aspect of the embodiments of this application, a method for determining the credit score of a loan user is provided, comprising: obtaining N types of financial data of the loan user with the user's authorization, wherein N is an integer greater than 1; calculating mathematical statistics of the i-th type of financial data when missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, wherein the i-th type of financial data is any one of the N types of financial data; correcting the missing values ​​and / or outliers in the i-th type of financial data according to the mathematical statistics, and extracting features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; and determining the target credit score of the loan user according to the N sets of feature sequences corresponding to the N types of financial data.

[0008] Optionally, if missing values ​​and / or outliers are detected in the i-th type of financial data among N types of financial data, mathematical statistics of the i-th type of financial data are calculated, including: if missing values ​​are detected in the i-th type of financial data, calculating the mean of the i-th type of financial data; and / or; if outliers are detected in the i-th type of financial data, calculating the mean and standard deviation of the i-th type of financial data.

[0009] Optionally, missing values ​​and / or outliers in the i-th type of financial data are corrected based on mathematical statistics, including: when missing values ​​are detected in the i-th type of financial data, filling the missing values ​​in the i-th type of financial data with the average value of the i-th type of financial data; and / or; when outliers are detected in the i-th type of financial data, calculating the difference between the average value of the i-th type of financial data and the outlier, and performing any of the following correction operations on the outlier based on the absolute value of the difference and the standard deviation of the i-th type of financial data: retaining the outlier, deleting the outlier, or modifying the outlier.

[0010] Optionally, outliers can be modified or deleted based on the absolute value of the difference and the standard deviation of the i-th type of financial data, including: calculating the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation; deleting or modifying outliers when the absolute value of the difference is detected to be greater than the first value, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value; and retaining outliers when the absolute value of the difference is detected to be less than or equal to the first value.

[0011] Optionally, the credit score of a loan user is determined based on N sets of feature sequences corresponding to N types of financial data, including: inputting the N sets of feature sequences into N model branches, wherein the j-th model branch is used to process the j-th set of feature sequences, and the j-th type of financial data corresponding to the j-th set of feature sequences is the same type of data as the training data of the j-th model branch; determining an initial credit score for the loan user by each model branch based on the input set of feature sequences and the prior knowledge of the model branch during training, wherein the N model branches generate N initial credit scores based on the N sets of feature sequences; and calculating the average of the N initial credit scores to obtain the target credit score for the loan user.

[0012] Optionally, the credit score of a loan user is determined based on N sets of feature sequences corresponding to N types of financial data, including: classifying the N sets of feature sequences into a first type of feature sequence and a second type of feature sequence, wherein the number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, and the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user; inputting the first type of feature sequence into a first prediction model, and determining a first credit score based on the first type of feature sequence using the prior knowledge learned by the first prediction model during the model training phase, wherein the first credit score is a value greater than 0, and the training data of the first prediction model is historical loan data. The system uses the credit behavior data of borrowers. The training label of the first prediction model is a credit score set based on the credit behavior data of historical loan users. The second type of feature sequence is input into the second prediction model. Based on the prior knowledge learned by the second prediction model during the model training phase, a second credit score is determined according to the second type of feature sequence. The second credit score is a value less than 0. The training data of the second prediction model is the dishonest behavior data of historical loan users. The training label of the second prediction model is a credit score set based on the dishonest behavior data of historical loan users. The first credit score and the second credit score are summed to obtain the target credit score of the loan user.

[0013] Optionally, the training process of the first prediction model includes the following steps: extracting L model training features from the credit behavior data of historical loan users, where L is an integer greater than 1; using a preset function to select at least one target training feature from the L model training features, where the preset function is used to determine the feature selection range of the L model training features; iteratively training the neural network based on the at least one target training feature until the neural network enters a convergent state, where the difference between the credit score predicted by the neural network based on the at least one target training feature in the convergent state and the credit score set for the credit behavior data of historical loan users is less than a preset threshold.

[0014] Furthermore, according to another aspect of the embodiments of this application, an apparatus for determining the credit score of a loan user is also provided, comprising: an acquisition unit, configured to acquire N types of financial data of the loan user with the user's authorization, wherein N is an integer greater than 1; a calculation unit, configured to calculate mathematical statistics of the i-th type of financial data when missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, wherein the i-th type of financial data is any one of the N types of financial data; a correction and extraction unit, configured to correct the missing values ​​and / or outliers in the i-th type of financial data according to the mathematical statistics, and to extract features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; and a determination unit, configured to determine the target credit score of the loan user according to the N sets of feature sequences corresponding to the N types of financial data.

[0015] Optionally, the calculation unit includes a first calculation subunit and a second calculation subunit. The first calculation subunit is used to calculate the average value of the i-th type of financial data when a missing value is detected in the i-th type of financial data; the second calculation subunit is used to calculate the average value and standard deviation of the i-th type of financial data when outliers are detected in the i-th type of financial data.

[0016] Optionally, the correction extraction unit includes a first detection subunit and a second detection subunit. The first detection subunit is used to fill the missing values ​​in the i-th type of financial data with the average value of the i-th type of financial data when a missing value is detected. The second detection subunit is used to calculate the difference between the average value of the i-th type of financial data and the outlier when an outlier is detected in the i-th type of financial data, and perform any one of the following correction operations on the outlier based on the absolute value of the difference and the standard deviation of the i-th type of financial data: retain the outlier, delete the outlier, or modify the outlier.

[0017] Optionally, the second detection subunit includes: a first calculation module, a first detection module, and a second detection module. The first calculation module calculates the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation. The first detection module deletes or modifies outliers when the absolute value of the detected difference is greater than the first value, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value. The second detection module retains outliers when the absolute value of the detected difference is less than or equal to the first value.

[0018] Optionally, the determining unit includes: a first input subunit, a first determining subunit, and a third calculation subunit. The first input subunit is used to input N sets of feature sequences into N model branches, where the j-th model branch processes the j-th set of feature sequences, and the j-th type of financial data corresponding to the j-th set of feature sequences is the same type of data as the training data of the j-th model branch. The first determining subunit is used to determine an initial credit score for the loan user based on the input set of feature sequences and the prior knowledge of the model branch during training, where the N model branches generate N initial credit scores based on the N sets of feature sequences. The third calculation subunit is used to calculate the average of the N initial credit scores to obtain the target credit score for the loan user.

[0019] Optionally, the determining unit includes: a first classification subunit, a second determining subunit, a third determining subunit, and a fourth calculation subunit. The first classification subunit is used to classify N sets of feature sequences into a first type of feature sequence and a second type of feature sequence. The number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, while the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user. The second determining subunit is used to input the first type of feature sequence into a first prediction model. Based on the prior knowledge learned during the model training phase, the first prediction model determines a first credit score according to the first type of feature sequence. The first credit score is a value greater than 0. The training data for the first prediction model is historical credit behavior data of loan users. The training label is a credit score set based on the credit behavior data of historical loan users; the third determination subunit is used to input the second type of feature sequence into the second prediction model, and determine the second credit score based on the second type of feature sequence using the prior knowledge learned by the second prediction model during the model training phase, wherein the second credit score is a value less than 0, the training data of the second prediction model is the data of dishonest behavior of historical loan users, and the training label of the second prediction model is a credit score set based on the data of dishonest behavior of historical loan users; the fourth calculation subunit is used to sum the first credit score and the second credit score to obtain the target credit score of the loan user.

[0020] Optionally, the training process of the first prediction model includes the following steps: extracting L model training features from the credit behavior data of historical loan users, where L is an integer greater than 1; using a preset function to select at least one target training feature from the L model training features, where the preset function is used to determine the feature selection range of the L model training features; iteratively training the neural network based on the at least one target training feature until the neural network enters a convergent state, where the difference between the credit score predicted by the neural network based on the at least one target training feature in the convergent state and the credit score set for the credit behavior data of historical loan users is less than a preset threshold.

[0021] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to execute the above-described method for determining the credit score of a loan user.

[0022] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the above-described method for determining the credit score of a loan user when it runs.

[0023] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions, which, when executed by a processor, implement the steps of the above-described method for determining the credit score of a loan user.

[0024] In this embodiment, firstly, with user authorization, N types of financial data of the loan user are obtained, where N is an integer greater than 1; if missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, mathematical statistics of the i-th type of financial data are calculated, where the i-th type of financial data is any one of the N types of financial data; the missing values ​​and / or outliers in the i-th type of financial data are corrected according to the mathematical statistics, and features are extracted from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; the target credit score of the loan user is determined according to the N sets of feature sequences corresponding to the N types of financial data.

[0025] As stated above, this application first ensures that N types of financial data from loan users are collected in strict compliance with privacy regulations and with the explicit authorization of the users, where N represents the number of data types and is an integer greater than 1. This step emphasizes the importance of user privacy protection while ensuring the legality and compliance of the data sources.

[0026] Next, this application conducts in-depth quality checks on the collected data, automatically identifying and handling missing and outlier values. By calculating mathematical statistics (such as mean and standard deviation) for each type of financial data, a high-quality data foundation is provided for subsequent feature extraction and model training. The automation and intelligence of data preprocessing significantly improve the efficiency and reliability of the evaluation process.

[0027] Secondly, for financial data with detected missing values ​​and / or outliers, this application employs advanced data correction techniques, such as mean imputation for missing values ​​and statistical correction for outliers, to ensure data integrity and accuracy. Subsequently, feature extraction is performed on the corrected data to generate a set of feature sequences closely related to the i-th type of financial data. This process effectively transforms the raw data into a form understandable to the model, providing direct input for credit scoring.

[0028] Finally, this application utilizes the extracted N sets of feature sequences to determine the target credit score for loan users through comprehensive analysis and algorithmic models. This score is not only based on single-dimensional data but also comprehensively considers the user's various financial situations, thereby yielding a more objective and accurate credit assessment result and providing a basis for decision-making for lending institutions.

[0029] In this embodiment of the application, by adopting the above method, through steps such as authorized data acquisition, data quality assurance, feature extraction and correction, and multi-dimensional feature analysis, an intelligent framework that can both protect user privacy and accurately assess credit status is constructed. This achieves the technical effect of optimizing data processing flow and improving the accuracy of user credit score determination, thereby solving the technical problem of low accuracy of user credit score determination caused by the large number of types of user data in the prior art. Attached Figure Description

[0030] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0031] Figure 1 A hardware block diagram of a computer terminal for implementing a method for determining the credit score of loan users is shown.

[0032] Figure 2 This is a flowchart of a method for determining the credit score of a loan user according to an embodiment of this application;

[0033] Figure 3 This is a simplified flowchart of a personal credit assessment based on the Bayesian optimization algorithm (SVM) according to an embodiment of this application.

[0034] Figure 4 This is an overall flowchart of an optimization based on SVM and Bayesian algorithms according to an embodiment of this application;

[0035] Figure 5 This is a schematic diagram of an optional device for determining the credit score of a loan user according to an embodiment of this application;

[0036] Figure 6 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0037] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0038] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0039] According to an embodiment of this application, a method embodiment for determining the credit score of a loan user is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0040] It should be noted that a credit scoring system can serve as the executing entity for the method of determining the credit score of a loan user in this application embodiment. It is understood that the method for determining the credit score of a loan user provided in this application embodiment can also be executed by other systems or devices, and this application embodiment does not specifically limit this.

[0041] The method embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing a method to determine the credit score of loan users is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0042] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits can be implemented wholly or partially as software, hardware, firmware, or any other combination. Furthermore, the data processing circuits can be a single, independent processing module, or wholly or partially integrated into any other element within the computer terminal 10 (or mobile device). As in the method for determining the credit score of a loan user as described in the embodiments of this application, the data processing circuit serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0043] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining the credit score of loan users in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned method for determining the credit score of loan users. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0044] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0045] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0046] Figure 2 This is a method for determining the credit score of loan users according to embodiments of this application, such as... Figure 2 As shown, the method includes the following steps:

[0047] Step S201: With the user's authorization, obtain N types of financial data of the loan user, where N is an integer greater than 1.

[0048] Optionally, with the user's authorization, customer loan-related data can be obtained from the financial institution's internal database.

[0049] Optionally, the loan user's N types of financial data include, but are not limited to, characteristics such as age x1, income x2, debt x3, number of overdue payments x4, and credit assessment data y (y=1 indicates good credit, y=-1 indicates poor credit).

[0050] Step S202: If missing values ​​and / or outliers are detected in the i-th type of financial data in the N types of financial data, calculate the mathematical statistics of the i-th type of financial data.

[0051] In step S202, the i-th type of financial data is any one of the N types of financial data.

[0052] Optionally, in the process of financial data collection and analysis, missing values ​​refer to situations where complete information cannot be obtained in the data records. This may be due to negligence in the data collection process, technical failures, implementation of data privacy protection measures, or other unforeseen reasons. The existence of missing values ​​can seriously affect the integrity of the data and the accuracy of the model.

[0053] Alternatively, in the process of financial data collection and analysis, outliers refer to values ​​in the dataset that deviate significantly from other data points. These values ​​may be caused by measurement errors, data entry errors, or extreme events. The existence of outliers may also distort the data analysis results, leading to poor model training performance.

[0054] Optionally, to ensure the quality of N types of financial data, the credit scoring system employs data inspection techniques. Data integrity checks can automatically identify whether missing values ​​exist in the i-th type of financial data. This process typically involves iterating through the data fields, checking each field for non-null values, thus ensuring that every part of the data is fully considered.

[0055] Optionally, for outlier detection, the credit scoring system uses statistical methods to identify values ​​that deviate significantly from the mean by calculating mathematical statistics such as the mean and standard deviation of the i-th type of financial data.

[0056] Optionally, after detecting missing or outlier values, the credit scoring system calculates mathematical statistics for the i-th type of financial data to support subsequent data correction. Key statistics include the mean and standard deviation.

[0057] Step S203: Correct the missing values ​​and / or outliers in the i-th type of financial data according to mathematical statistics, and extract features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data.

[0058] Optionally, for correcting missing values, the credit scoring system employs a dynamic and refined approach, selecting the most appropriate imputation strategy based on the characteristics of different types of financial data. For basic information data of loan users (such as age and income), if missing values ​​exist, the credit scoring system first calculates statistical measures such as the mean, median, or mode of this type of data, and then selects the most suitable imputation method based on the data's distribution characteristics. For normally distributed data, the credit scoring system uses the mean for imputation; while for non-normally distributed data or data with extreme values, the credit scoring system uses the median or interpolation based on the relationships between data points to more accurately reflect the true distribution of the data, thereby avoiding model bias caused by improper imputation of missing values.

[0059] Optionally, the credit scoring system employs a rigorous strategy when handling outliers. First, it calculates the relevant statistical measures of the outliers to identify values ​​that significantly deviate from the central tendency of the dataset. Then, based on the nature of the outliers and their impact on model training, the credit scoring system dynamically decides whether to retain, delete, or correct them.

[0060] Optionally, feature extraction is a crucial step in training machine learning models, aiming to select the most influential subset of features for model prediction from raw, multi-dimensional financial data. In this application, feature extraction encompasses not only simple data transformation and dimensionality reduction but also feature engineering techniques such as feature construction, feature selection, and feature encoding, in order to obtain a set of feature sequences that clearly reflect the creditworthiness of loan users.

[0061] Step S204: Determine the target credit score for loan users based on the N sets of feature sequences corresponding to the N types of financial data.

[0062] Optionally, determining the target credit score for a loan user starts with the construction of feature vectors, and through key steps such as model training and parameter optimization, model validation and testing, a comprehensive and accurate credit score is finally generated.

[0063] Optionally, this application uses the Support Vector Machine (SVM) algorithm as the core model and combines it with Bayesian optimization to intelligently adjust the model parameters. The model will predict the user's credit status based on user characteristics and convert it into a quantitative credit score.

[0064] Through the above steps, in this embodiment of the application, firstly, with the user's authorization, N types of financial data of the loan user are obtained, where N is an integer greater than 1; if missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, mathematical statistics of the i-th type of financial data are calculated, where the i-th type of financial data is any one of the N types of financial data; the missing values ​​and / or outliers in the i-th type of financial data are corrected according to the mathematical statistics, and features are extracted from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; the target credit score of the loan user is determined according to the N sets of feature sequences corresponding to the N types of financial data.

[0065] As stated above, this application first ensures that N types of financial data from loan users are collected in strict compliance with privacy regulations and with the explicit authorization of the users, where N represents the number of data types and is an integer greater than 1. This step emphasizes the importance of user privacy protection while ensuring the legality and compliance of the data sources.

[0066] Next, this application conducts in-depth quality checks on the collected data, automatically identifying and handling missing and outlier values. By calculating mathematical statistics (such as mean and standard deviation) for each type of financial data, a high-quality data foundation is provided for subsequent feature extraction and model training. The automation and intelligence of data preprocessing significantly improve the efficiency and reliability of the evaluation process.

[0067] Secondly, for financial data with detected missing values ​​and / or outliers, this application employs advanced data correction techniques, such as mean imputation for missing values ​​and statistical correction for outliers, to ensure data integrity and accuracy. Subsequently, feature extraction is performed on the corrected data to generate a set of feature sequences closely related to the i-th type of financial data. This process effectively transforms the raw data into a form understandable to the model, providing direct input for credit scoring.

[0068] Finally, this application utilizes the extracted N sets of feature sequences to determine the target credit score for loan users through comprehensive analysis and algorithmic models. This score is not only based on single-dimensional data but also comprehensively considers the user's various financial situations, thereby yielding a more objective and accurate credit assessment result and providing a basis for decision-making for lending institutions.

[0069] In this embodiment of the application, by adopting the above method, through steps such as authorized data acquisition, data quality assurance, feature extraction and correction, and multi-dimensional feature analysis, an intelligent framework that can both protect user privacy and accurately assess credit status is constructed. This achieves the technical effect of optimizing data processing flow and improving the accuracy of user credit score determination, thereby solving the technical problem of low accuracy of user credit score determination caused by the large number of types of user data in the prior art.

[0070] In one alternative embodiment, the credit scoring system first calculates the average of the i-th type of financial data if a missing value is detected in the i-th type of financial data, and secondly, and / or, calculates the average and standard deviation of the i-th type of financial data if an outlier is detected in the i-th type of financial data.

[0071] Optionally, the collected data can be preprocessed, such as cleansing, standardization, and missing value imputation, to improve data quality.

[0072] Alternatively, the mean (also known as the arithmetic mean) is the sum of all values ​​in a dataset divided by the number of data points. It is one of the most basic statistics describing the trend in a dataset.

[0073] Optionally, when the credit scoring system detects a missing value in the i-th type of financial data, such as a missing x2 (income), let its corresponding data vector be x. i =(x i1 x i2 x i3 x i4 If the mean is used for filling, let the set of all income values ​​be... (where n is the number of data points), then the fill value is (i.e., the average value of the i-th type of financial data).

[0074] Optionally, standard deviation is an important statistic for measuring the dispersion of data distribution. A larger standard deviation indicates greater differences between data points and a more dispersed data distribution; conversely, a smaller standard deviation indicates more concentrated data points and a more uniform distribution. In financial data processing, standard deviation is often used to detect outliers and help assess the stability of the data and the reliability of predictive models.

[0075] Optionally, when the credit scoring system detects outliers in the i-th type of financial data, it calculates the mean and standard deviation of the i-th type of financial data. The mean can be calculated using... The standard deviation can be calculated using... calculate.

[0076] As can be seen from the above, for methods of handling missing and outlier values, calculating the mean of missing values, as well as the mean and standard deviation of outliers, can improve the quality of data and provide a solid foundation for subsequent feature extraction, model training, and credit scoring.

[0077] In one optional embodiment, the credit scoring system first fills the missing values ​​in the i-th type of financial data with the average value of the i-th type of financial data if missing values ​​are detected. Secondly, and / or, if outliers are detected in the i-th type of financial data, the system calculates the difference between the average value of the i-th type of financial data and the outliers, and performs any of the following correction operations on the outliers based on the absolute value of the difference and the standard deviation of the i-th type of financial data: retaining the outliers, deleting the outliers, or modifying the outliers.

[0078] Optionally, when the credit scoring system detects a missing value in the i-th type of financial data, such as a missing x2 (income), let its corresponding data vector be x. i =(x i1 x i2 x i3 x i4 If the mean is used for filling, let the set of all income values ​​be... (where n is the number of data points), then the fill value is (i.e., the average value of the i-th type of financial data).

[0079] Optionally, when the credit scoring system detects an outlier in the i-th type of financial data, it calculates the difference between the average of the i-th type of financial data and the outlier; that is, if the outlier x... 2k The data follows a normal distribution, with a mean of μ and a standard deviation of σ. Then, for data point x... 2k Calculate |x 2k -μ|.

[0080] Optionally, based on the absolute value of the difference between the mean and outliers of the i-th type of financial data and the standard deviation of the i-th type of financial data, i.e., based on |x 2k The values ​​of -μ| and σ determine which of the following correction operations can be performed on outliers: retain outliers, delete outliers, or modify outliers.

[0081] As described above, for missing values, the credit scoring system employs a feature-based average imputation strategy. This method fully utilizes existing information within the dataset, imputing missing values ​​by calculating the average of non-missing data. This effectively avoids prediction bias caused by incomplete data and ensures that the input data for model training is complete and representative. For outliers, intelligent processing is performed based on the absolute value of the difference between the average of the i-th type of financial data and the outlier, and the standard deviation of that type of data. This eliminates the influence of outliers and ensures data quality and consistency, providing a high-quality data foundation for subsequent feature extraction, model training, and credit score determination.

[0082] In one optional embodiment, the credit scoring system first calculates the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation. Then, if the absolute value of the difference is detected to be greater than the first value, outliers are deleted or modified, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value. Finally, if the absolute value of the difference is detected to be less than or equal to the first value, the outliers are retained.

[0083] Optionally, the preset coefficient is a manually defined value used to amplify or reduce the standard deviation.

[0084] Optionally, the first value refers to the anomaly index calculated based on the absolute value of the difference between the outlier and the mean of the data in its category, as well as the standard deviation of that category of data. The calculation of the first value not only quantifies the degree of deviation of the outlier but also takes into account the inherent volatility within the dataset, providing an objective basis for determining the outlier handling strategy.

[0085] Optionally, for outliers whose first value significantly exceeds a preset threshold, the credit scoring system adopts a strategy of direct deletion. This is because such outliers often represent data entry errors, measurement mistakes, or atypical events. The presence of these outliers can severely distort the model's training results, causing the model to make inaccurate predictions when faced with normal data. By deleting these significantly abnormal values, the purity and representativeness of the model training data can be ensured, avoiding the adverse effects of outliers on model training.

[0086] Optionally, for outliers within a certain range, the credit scoring system employs a modification strategy. Specifically, based on the difference between the outlier and the average of its category, combined with the standard deviation of that category, the credit scoring system implements a reasonable numerical adjustment. This strategy aims to preserve potentially valuable information within outliers while minimizing their negative impact on model performance. By intelligently correcting rather than simply eliminating them, it preserves data diversity and enhances the model's adaptability to complex financial environments.

[0087] As can be seen from the above, the strategy of deleting and modifying outliers based on the absolute value of the difference between outliers and the average value of the data in their respective categories, as well as the anomaly index calculated from the standard deviation of that category of data, can improve data quality and enhance the generalization ability of the model.

[0088] In one optional embodiment, the credit scoring system first inputs N sets of feature sequences into N model branches, where the j-th model branch processes the j-th set of feature sequences. The j-th type of financial data corresponding to the j-th set of feature sequences is the same type of data as the training data of the j-th model branch. Then, each model branch determines an initial credit score for the loan user based on the input set of feature sequences and the prior knowledge of the model branch during training. The N model branches generate N initial credit scores based on the N sets of feature sequences. Finally, the average of the N initial credit scores is calculated to obtain the target credit score for the loan user.

[0089] Optionally, the credit scoring system employs a multi-model branch fusion approach, inputting N carefully selected feature sequences into N independent model branches for evaluation. Each feature sequence focuses on different aspects of individual loan credit assessment, such as income stability, debt ratio, credit history, and asset status, thereby constructing a comprehensive and in-depth user credit profile.

[0090] Optionally, in each model branch, the credit scoring system uses the corresponding feature sequence to perform a preliminary credit score on the loan user. This step aims to independently assess the user's creditworthiness from multiple perspectives. Each initial credit score is based on the prediction results of the specific branch model, reflecting the user's creditworthiness under that feature sequence.

[0091] Optionally, the credit scoring system calculates the average of N initial credit scores as a preliminary result of the user's credit score, integrates the scoring information from N model branches, and balances the evaluation results of different model branches.

[0092] As described above, the credit scoring system inputs N sets of feature sequences into N model branches, enabling in-depth analysis of the impact of each feature on the credit score and avoiding mutual interference between features. Then, an averaging operation balances the evaluation results of different model branches, avoiding biases or errors that might arise from a single model, thus providing users with a more comprehensive and objective credit score.

[0093] In one optional embodiment, the credit scoring system first classifies N sets of feature sequences into a first type of feature sequence and a second type of feature sequence. The number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, while the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user. Next, the first type of feature sequence is input into a first prediction model. Based on the prior knowledge learned during the model training phase, the first credit score is determined using the first type of feature sequence. The first credit score is a value greater than 0. The training data for the first prediction model consists of historical loan user integrity behavior data, and the training label for the first prediction model is a credit score set based on the historical loan user integrity behavior data. Then, the second type of feature sequence is input into a second prediction model. Based on the prior knowledge learned during the model training phase, the second credit score is determined using the second type of feature sequence. The second credit score is a value less than 0. The training data for the second prediction model consists of historical loan user dishonesty behavior data, and the training label for the second prediction model is a credit score set based on the historical loan user dishonesty behavior data. Finally, the first credit score and the second credit score are summed to obtain the loan user's target credit score.

[0094] Optionally, the credit scoring system further subdivides the N sets of feature sequences into a first category and a second category of feature sequences based on their different contributions and relevance to credit assessment.

[0095] Optionally, the number of features in the first type of feature sequence is positively correlated with the credit score of the loan user; while the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user.

[0096] Optionally, the first prediction model focuses on processing features that directly reflect credit level and can quickly and accurately provide assessment results based on core credit indicators; while the second prediction model focuses more on assessing the impact of background information on credit and further improves the user's credit profile by introducing indirect features.

[0097] Optionally, the credit scoring system sums the outputs of the first and second prediction models—namely, the first credit score and the second credit score—to calculate the final credit score for the loan user. This calculation step reflects the application's concept of multi-level and multi-faceted assessment of the user's creditworthiness. Through the summation operation, it can comprehensively consider the influence of direct and indirect credit factors to obtain a more comprehensive and balanced credit score result.

[0098] As described above, the credit scoring system categorizes N sets of feature sequences into first-class and second-class feature sequences based on their varying contributions and relevance to credit assessment. This facilitates a deeper understanding of the independent impact of different feature sequences on credit scoring. Then, the first and second prediction models are used for evaluation, and the first and second credit scores are summed to construct a hierarchical, comprehensive, and accurate credit assessment system. This significantly improves the efficiency and accuracy of banks' decision-making in personal loan credit risk management.

[0099] In one optional embodiment, the credit scoring system first extracts L model training features from the credit behavior data of historical loan users, where L is an integer greater than 1. Then, it uses a preset function to filter at least one target training feature from the L model training features, where the preset function is used to determine the feature filtering range of the L model training features. Finally, it iteratively trains the neural network based on the at least one target training feature until the neural network enters a convergent state. The difference between the credit score predicted by the neural network based on the at least one target training feature in the convergent state and the credit score set for the credit behavior data of historical loan users is less than a preset threshold.

[0100] Optionally, the dataset used for model training needs to be partitioned. Let the dataset be D = {(x1, y1), (x2, y2), ..., (x...} n y n Divide it into training set D train Validation set D val and test set D test Satisfy |D train |=0.6n,|D val |=0.2n,|D test |=0.2n.

[0101] Optionally, the credit scoring system extracts L model training features from historical loan user credit behavior data. These features cover multiple dimensions of information, including the borrower's financial situation, repayment history, and credit record, aiming to construct a feature space that comprehensively reflects an individual's credit risk.

[0102] Optionally, the preset function selects at least one target training feature from L model training features. If the radial basis function (RBF) is chosen as the preset function, its formula is K(x i x j )=exp(-γ||x i -x j || 2 ), where γ is the kernel parameter (where x i and x j (These represent two different training features).

[0103] Optionally, the credit scoring system needs to set an initial parameter range, mainly including setting the initial search range of the penalty parameter C to [C]. min C max The initial search range for the kernel parameter γ is set to [γ]. min γ max ].

[0104] Optionally, a maximum number of iterations can be set. When the number of iterations reaches the maximum, the algorithm iteration stops, and the optimal sampling point and the corresponding optimal value of the objective function are output, which are the optimal parameters C and γ of the SVM model.

[0105] Optionally, the F1 score and AUC value are introduced as evaluation criteria for the model's prediction performance to evaluate the overall prediction performance of the model.

[0106] The formula for calculating the f1 evaluation index is as follows:

[0107]

[0108] Recall and precision are called the detection rate and the accuracy, respectively.

[0109]

[0110] In this model, TP represents the number of positive samples predicted as positive; FN represents the number of positive samples predicted as negative; and FP represents the number of negative samples predicted as positive. A higher recall indicates a stronger ability to identify positive samples, while a higher precision indicates a stronger ability to distinguish negative samples. F1 is a combination of both; a higher F1 indicates a more robust classification model. Decreasing either recall or precision will decrease the value of F1.

[0111] The AUC value is calculated as follows:

[0112]

[0113] The AUC (Area Under Curve) value is the area enclosed by the ROC (Receiver Operating Characteristic) curve and the coordinate system. The horizontal axis represents the FPR (False Positive Rate), which indicates the probability that the actual value is negative and the prediction is positive. The vertical axis represents the TPR (True Positive Rate), which indicates the probability that the actual value is positive and the prediction is also positive.

[0114] As can be seen from the above, the credit scoring system intelligently extracts and filters key features from historical loan user credit behavior data, and combines this with iterative training of deep learning neural networks to build an efficient and accurate personal loan credit assessment system. This improves the model's predictive performance and provides more reliable data support for banks' personal loan decisions.

[0115] In one alternative embodiment, Figure 3 This application illustrates a personal credit assessment method based on the Bayesian optimization algorithm (SVM) according to an embodiment of the present application. Figure 3 As shown, the process begins with data collection and preprocessing. Customer loan-related data is retrieved from an internal database, and then the collected data undergoes cleaning, standardization, and missing value imputation to improve data quality. Next, an SVM model is trained, selecting an appropriate kernel function and parameters. Then, a Bayesian model is trained, and the Gaussian process surrogate model is updated for optimization. The process is iterated to output the optimal parameters C and γ of the SVM model. Finally, credit assessment criteria are set to obtain a credit score.

[0116] In one alternative embodiment, Figure 4 This illustrates an overall process for optimization based on SVM and Bayesian algorithms according to an embodiment of this application, such as... Figure 4 As shown:

[0117] 1. Collect multi-dimensional information about customers, including basic information, financial information, credit history, and behavioral habits;

[0118] 2. Perform preprocessing operations such as cleaning, standardization, and missing value imputation on the collected data to improve data quality;

[0119] 3. Select key features related to credit risk from the preprocessed data.

[0120] 4. By using kernel functions to map the data from the original feature space to a high-dimensional feature space, the linear inner product operation is made nonlinear. Then, an optimal hyperplane that maximizes the classification margin is established in the high-dimensional feature space.

[0121] 5. Find the parameters w and b that maximize γ.

[0122] 6. Randomly select n0 sampling points within the search range set by C and γ in the SVM model. Use the average test accuracy of ten-fold cross-validation as the objective function f. Use different combinations of model parameters as independent variables x to form a surrogate model framework, and obtain the initial distribution of the objective function and the sampling point set D1.

[0123] 7. Select the next sampling point x by maximizing the acquisition function. t The sampling point function f(x) is obtained. t ).

[0124] 8. Transfer the new sampling point [x] t f(x) t Add the sample point set D1 to update the Gaussian process surrogate model, making the surrogate model more closely match the distribution of the objective function.

[0125] 9. When the number of iterations reaches the maximum, stop the algorithm iteration and output the optimal sampling point and the corresponding optimal value of the objective function.

[0126] 10. Introduce the f1 evaluation index as the evaluation standard for the model.

[0127] 11. Introduce AUC value to evaluate the overall prediction performance of the model.

[0128] According to another aspect of the embodiments of this application, a device for determining the credit score of a loan user is also provided, wherein, Figure 5 This is a schematic diagram of an optional loan user credit score determination device according to an embodiment of this application, as shown below. Figure 5 As shown, the device for determining the credit score of a loan user includes: an acquisition unit 501, a calculation unit 502, a correction extraction unit 503, and a determination unit 504.

[0129] The acquisition unit 501 is used to acquire N types of financial data of the loan user with the user's authorization, where N is an integer greater than 1; the calculation unit 502 is used to calculate mathematical statistics of the i-th type of financial data when missing values ​​and / or outliers are detected in the i-th type of financial data in the N types of financial data, where the i-th type of financial data is any one of the N types of financial data; the correction and extraction unit 503 is used to correct the missing values ​​and / or outliers in the i-th type of financial data according to the mathematical statistics, and to extract features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; the determination unit 504 is used to determine the target credit score of the loan user according to the N sets of feature sequences corresponding to the N types of financial data.

[0130] Optionally, the calculation unit 501 includes a first calculation subunit and a second calculation subunit. The first calculation subunit is used to calculate the average value of the i-th type of financial data when a missing value is detected in the i-th type of financial data; the second calculation subunit is used to calculate the average value and standard deviation of the i-th type of financial data when outliers are detected in the i-th type of financial data.

[0131] Optionally, the correction extraction unit 503 includes a first detection subunit and a second detection subunit. The first detection subunit is used to fill the missing values ​​in the i-th type of financial data with the average value of the i-th type of financial data when a missing value is detected. The second detection subunit is used to calculate the difference between the average value of the i-th type of financial data and the outlier when an outlier is detected in the i-th type of financial data, and perform any one of the following correction operations on the outlier based on the absolute value of the difference and the standard deviation of the i-th type of financial data: retain the outlier, delete the outlier, or modify the outlier.

[0132] Optionally, the second detection subunit includes: a first calculation module, a first detection module, and a second detection module. The first calculation module calculates the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation. The first detection module deletes or modifies outliers when the absolute value of the detected difference is greater than the first value, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value. The second detection module retains outliers when the absolute value of the detected difference is less than or equal to the first value.

[0133] Optionally, the determining unit 504 includes: a first input subunit, a first determining subunit, and a third calculation subunit. The first input subunit is used to input N sets of feature sequences into N model branches, where the j-th model branch processes the j-th set of feature sequences, and the j-th type of financial data corresponding to the j-th set of feature sequences is the same type of data as the training data of the j-th model branch. The first determining subunit is used to determine an initial credit score for the loan user based on the input set of feature sequences and the prior knowledge of the model branch during training, where the N model branches generate N initial credit scores based on the N sets of feature sequences. The third calculation subunit is used to calculate the average of the N initial credit scores to obtain the target credit score for the loan user.

[0134] Optionally, the determining unit 504 includes: a first classification subunit, a second determining subunit, a third determining subunit, and a fourth calculation subunit. The first classification subunit is used to classify N sets of feature sequences into a first type of feature sequence and a second type of feature sequence. The number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, while the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user. The second determining subunit is used to input the first type of feature sequence into a first prediction model, and determine a first credit score based on the first type of feature sequence using prior knowledge learned during the model training phase. The first credit score is a value greater than 0. The training data for the first prediction model is historical credit behavior data of loan users. The training label is a credit score set based on the credit behavior data of historical loan users; the third determination subunit is used to input the second type of feature sequence into the second prediction model, and determine the second credit score based on the second type of feature sequence using the prior knowledge learned by the second prediction model during the model training phase, wherein the second credit score is a value less than 0, the training data of the second prediction model is the data of dishonest behavior of historical loan users, and the training label of the second prediction model is a credit score set based on the data of dishonest behavior of historical loan users; the fourth calculation subunit is used to sum the first credit score and the second credit score to obtain the target credit score of the loan user.

[0135] Optionally, the training process of the first prediction model includes the following steps: extracting L model training features from the credit behavior data of historical loan users, where L is an integer greater than 1; using a preset function to select at least one target training feature from the L model training features, where the preset function is used to determine the feature selection range of the L model training features; iteratively training the neural network based on the at least one target training feature until the neural network enters a convergent state, where the difference between the credit score predicted by the neural network based on the at least one target training feature in the convergent state and the credit score set for the credit behavior data of historical loan users is less than a preset threshold.

[0136] It should be noted that the above-mentioned modules or units may be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above-mentioned modules may also be part of the device and may run in the computer terminal 10 provided in the embodiment.

[0137] Embodiments of this application may provide an electronic device. Figure 6 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 6As shown, the electronic device may include: one or more ( Figure 6 Only one is shown in the diagram. The processor, memory, memory controller, and peripheral interface are connected to the radio frequency module, audio module, and display.

[0138] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0139] The processor can access information and applications stored in memory via a transmission device to perform the following steps: With user authorization, acquire N types of financial data of the loan user, where N is an integer greater than 1; if missing values ​​and / or outliers are detected in the i-th type of financial data among the N types, calculate mathematical statistics for the i-th type of financial data, where the i-th type of financial data is any one of the N types; correct the missing values ​​and / or outliers in the i-th type of financial data based on the mathematical statistics, and extract features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data; determine the target credit score of the loan user based on the N sets of feature sequences corresponding to the N types of financial data.

[0140] The processor can invoke information and applications stored in memory via a transmission device to perform the following steps: if a missing value is detected in the i-th type of financial data, calculate the average of the i-th type of financial data; and / or, if an outlier is detected in the i-th type of financial data, calculate the average and standard deviation of the i-th type of financial data.

[0141] The processor can invoke information and application programs stored in the memory via a transmission device to perform the following steps: if a missing value is detected in the i-th type of financial data, fill the missing value in the i-th type of financial data with the average value of the i-th type of financial data; and / or; if an outlier is detected in the i-th type of financial data, calculate the difference between the average value of the i-th type of financial data and the outlier, and perform any of the following correction operations on the outlier based on the absolute value of the difference and the standard deviation of the i-th type of financial data: retain the outlier, delete the outlier, or modify the outlier.

[0142] The processor can access the information and application program stored in the memory via the transmission device to perform the following steps: calculate the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation; if the absolute value of the detected difference is greater than the first value, delete or modify the outlier, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value; if the absolute value of the detected difference is less than or equal to the first value, retain the outlier.

[0143] The processor can access information and applications stored in memory via a transmission device to perform the following steps: Inputting N sets of feature sequences into N model branches, where the j-th model branch processes the j-th set of feature sequences, and the j-th type of financial data corresponding to the j-th set of feature sequences is the same type of data as the training data of the j-th model branch; determining an initial credit score for the loan user based on the input set of feature sequences and the prior knowledge gained during the training process, where the N model branches generate N initial credit scores based on the N sets of feature sequences; calculating the average of the N initial credit scores to obtain the target credit score for the loan user.

[0144] The processor can access information and applications stored in memory via a transmission device to perform the following steps: classifying N sets of feature sequences into a first type of feature sequence and a second type of feature sequence, wherein the number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, and the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user; inputting the first type of feature sequence into a first prediction model, and determining a first credit score based on the first type of feature sequence using prior knowledge learned during the model training phase of the first prediction model, wherein the first credit score is a value greater than 0, and the training data of the first prediction model is historical loan data. The system uses the credit behavior data of borrowers. The training label of the first prediction model is a credit score set based on the credit behavior data of historical loan users. The second type of feature sequence is input into the second prediction model. Based on the prior knowledge learned by the second prediction model during the model training phase, a second credit score is determined according to the second type of feature sequence. The second credit score is a value less than 0. The training data of the second prediction model is the dishonest behavior data of historical loan users. The training label of the second prediction model is a credit score set based on the dishonest behavior data of historical loan users. The first credit score and the second credit score are summed to obtain the target credit score of the loan user.

[0145] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: extracting L model training features from the credit behavior data of historical loan users, where L is an integer greater than 1; using a preset function to filter at least one target training feature from the L model training features, where the preset function is used to determine the feature filtering range of the L model training features; iteratively training the neural network based on the at least one target training feature until the neural network enters a convergent state, where the difference between the credit score predicted by the neural network based on the at least one target training feature in the convergent state and the credit score set for the credit behavior data of historical loan users is less than a preset threshold.

[0146] In this embodiment, the first step is to ensure that N types of financial data from loan users are collected in strict compliance with privacy regulations and with explicit user authorization, where N represents the number of data types and is an integer greater than 1. This step emphasizes the importance of user privacy protection while ensuring the legality and compliance of the data sources.

[0147] Next, this application conducts in-depth quality checks on the collected data, automatically identifying and handling missing and outlier values. By calculating mathematical statistics (such as mean and standard deviation) for each type of financial data, a high-quality data foundation is provided for subsequent feature extraction and model training. The automation and intelligence of data preprocessing significantly improve the efficiency and reliability of the evaluation process.

[0148] Secondly, for financial data with detected missing values ​​and / or outliers, this application employs advanced data correction techniques, such as mean imputation for missing values ​​and statistical correction for outliers, to ensure data integrity and accuracy. Subsequently, feature extraction is performed on the corrected data to generate a set of feature sequences closely related to the i-th type of financial data. This process effectively transforms the raw data into a form understandable to the model, providing direct input for credit scoring.

[0149] Finally, this application utilizes the extracted N sets of feature sequences to determine the target credit score for loan users through comprehensive analysis and algorithmic models. This score is not only based on single-dimensional data but also comprehensively considers the user's various financial situations, thereby yielding a more objective and accurate credit assessment result and providing a basis for decision-making for lending institutions.

[0150] In this embodiment of the application, by adopting the above method, through steps such as authorized data acquisition, data quality assurance, feature extraction and correction, and multi-dimensional feature analysis, an intelligent framework that can both protect user privacy and accurately assess credit status is constructed. This achieves the technical effect of optimizing data processing flow and improving the accuracy of user credit score determination, thereby solving the technical problem of low accuracy of user credit score determination caused by the large number of types of user data in the prior art.

[0151] Those skilled in the art will understand that Figure 6 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 6 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 6 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 6 The different configurations shown.

[0152] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0153] According to another aspect of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, wherein when the computer program is executed, the device in which the computer-readable storage medium is located performs the above-described method for determining the credit score of a loan user.

[0154] According to another aspect of this application, a computer program product is also provided, wherein the computer program product includes computer instructions, wherein when the computer instructions are executed, the device on which the computer program product is located executes the aforementioned method for determining the credit score of a loan user.

[0155] The above-described embodiments or examples disclosed in this application are not exhaustive, but merely illustrative of some embodiments or examples, and are not intended to limit the scope of protection of this application. Unless otherwise specified, each step in a particular embodiment or example can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment or example can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment or example can be arbitrarily interchanged. Furthermore, optional methods or examples in a particular embodiment or example can be arbitrarily combined; moreover, various embodiments or examples can be arbitrarily combined. For example, some or all steps of different embodiments or examples can be arbitrarily combined, and a particular embodiment or example can be arbitrarily combined with optional methods or examples of other embodiments or examples.

[0156] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0157] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0158] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0159] The units described as separate components may or may not be physically separate. Similarly, the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0160] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0161] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0162] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the credit score of a loan user, characterized in that, include: With the user's authorization, obtain N types of financial data of the loan user, where N is an integer greater than 1; If missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, calculate the mathematical statistics of the i-th type of financial data, wherein the i-th type of financial data is any one of the N types of financial data; Based on the mathematical statistics, the missing values ​​and / or outliers in the i-th type of financial data are corrected, and features are extracted from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data. The target credit score of the loan user is determined based on the N sets of feature sequences corresponding to the N types of financial data. Specifically, when missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, the mathematical statistics of the i-th type of financial data are calculated, including: when missing values ​​are detected in the i-th type of financial data, calculating the average value of the i-th type of financial data; and / or: when outliers are detected in the i-th type of financial data, calculating the average value and standard deviation of the i-th type of financial data; The method for determining the target credit score of the loan user based on N sets of feature sequences corresponding to the N types of financial data includes: classifying the N sets of feature sequences into a first type of feature sequence and a second type of feature sequence, wherein the number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, and the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user; inputting the N sets of feature sequences into N model branches respectively, inputting the first type of feature sequence into a first prediction model, and inputting the second type of feature sequence into a second prediction model, wherein... The j-th model branch among the N model branches is used to process the j-th feature sequence among the N feature sequences. The j-th type of financial data corresponding to the j-th feature sequence is the same type of data as the training data of the j-th model branch. Each model branch determines an initial credit score for the loan user based on the input feature sequence and the prior knowledge of the model branch during training. The N model branches generate N initial credit scores based on the N feature sequences. The average of the N initial credit scores is calculated to obtain the target credit score for the loan user.

2. The method for determining the credit score of a loan user according to claim 1, characterized in that, Correcting missing and / or outlier values ​​in the i-th type of financial data based on the mathematical statistics includes: If a missing value is detected in the i-th type of financial data, the missing value in the i-th type of financial data is filled with the average value of the i-th type of financial data; and / or; If an outlier is detected in the i-th type of financial data, the difference between the average of the i-th type of financial data and the outlier is calculated, and the outlier is corrected using any of the following operations based on the absolute value of the difference and the standard deviation of the i-th type of financial data: Retain the outlier, delete the outlier, or modify the outlier.

3. The method for determining the credit score of a loan user according to claim 2, characterized in that, The outlier is corrected based on the absolute value of the difference and the standard deviation of the i-th type of financial data, including: Calculate the product of the standard deviation and a preset coefficient to obtain a first value, wherein the preset coefficient is used to amplify or reduce the standard deviation; If the absolute value of the difference is detected to be greater than the first value, the outlier is deleted or the outlier is modified, wherein the absolute value of the difference between the modified outlier and the average value of the i-th type of financial data is less than or equal to the first value; If the absolute value of the difference is detected to be less than or equal to the first value, the outlier value is retained.

4. The method for determining the credit score of a loan user according to claim 1, characterized in that, The target credit score for the loan user is determined based on N sets of feature sequences corresponding to the N types of financial data, including: The N sets of feature sequences are classified into a first type of feature sequence and a second type of feature sequence. The number of features in the first type of feature sequence is positively correlated with the credit score of the loan user, while the number of features in the second type of feature sequence is negatively correlated with the credit score of the loan user. The first type of feature sequence is input into the first prediction model. The first credit score is determined based on the first type of feature sequence using the prior knowledge learned by the first prediction model during the model training phase. The first credit score is a value greater than 0. The training data of the first prediction model is the credit behavior data of historical loan users. The training label of the first prediction model is the credit score set for the credit behavior data of the historical loan users. The second type of feature sequence is input into the second prediction model. The second credit score is determined based on the second type of feature sequence using the prior knowledge learned by the second prediction model during the model training phase. The second credit score is a value less than 0. The training data of the second prediction model is the data on the dishonest behavior of the historical loan users. The training label of the second prediction model is the credit score set for the data on the dishonest behavior of the historical loan users. The target credit score of the loan user is obtained by summing the first credit score and the second credit score.

5. The method for determining the credit score of a loan user according to claim 4, characterized in that, The training process of the first prediction model includes the following steps: L model training features are extracted from the credit behavior data of the historical loan users, where L is an integer greater than 1; At least one target training feature is obtained by filtering from the L model training features using a preset function, wherein the preset function is used to determine the feature filtering range of the L model training features; The neural network is iteratively trained based on the at least one target training feature until the neural network enters a convergent state, wherein the difference between the credit score predicted by the neural network in the convergent state based on the at least one target training feature and the credit score set for the credit behavior data of the historical loan user is less than a preset threshold.

6. An apparatus for determining the credit score of a loan user, used to implement the method for determining the credit score of a loan user as described in any one of claims 1 to 5, characterized in that, include: The acquisition unit is used to acquire N types of financial data of the loan user with the user's authorization, where N is an integer greater than 1; The calculation unit is configured to calculate mathematical statistics of the i-th type of financial data when missing values ​​and / or outliers are detected in the i-th type of financial data among the N types of financial data, wherein the i-th type of financial data is any one of the N types of financial data. The feature extraction unit is used to correct missing values ​​and / or outliers in the i-th type of financial data according to the mathematical statistics, and to extract features from the corrected i-th type of financial data to obtain a set of feature sequences corresponding to the i-th type of financial data. The scoring determination unit is used to determine the target credit score of the loan user based on the N sets of feature sequences corresponding to the N types of financial data.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the method for determining the credit score of a loan user as described in any one of claims 1 to 5.

8. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein, when the program runs, the processor executes the method for determining the credit score of a loan user as described in any one of claims 1 to 5.

9. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the method for determining the credit score of a loan user as described in any one of claims 1 to 5.

Images