A consumer finance credit investigation management system
Patent Information
- Application Number
- CN202411884313.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2026-10-23
AI Technical Summary
[0004]本发明的目的在于提供一种消费金融征信管理系统，以解决传统主要依赖于线下渠道和有限的金融机构数据源，征信数据覆盖不全面，难以全面反映消费者的信用状况的问题
[0034]本发明通过收集消费者多个数据源的信用信息并整合处理，可以全面了解消费者的信用情况，为后续分析提供数据基础，然后对整合后的数据进行大数据处理，实现数据的实时收集和处理，及时发现潜在的信用风险，然后使用建立好的信用评分模型对消费者的信用进行评分，预测其未来的信用表现，判断其信用风险水平，然后根据信用评分结果制定相应的风险控制策略，便于金融机构进行征信管理。
Smart Images
Figure CN122966352A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of financial management, specifically relating to a consumer finance credit reporting management system. Background Technology
[0002] Consumer finance credit reporting refers to the process of assessing, recording, and managing the credit status of individuals or businesses in the consumer finance sector by relying on credit information service agencies, financial institutions, and the collection and analysis of relevant data. Its main purpose is to help financial institutions determine the credit risk of borrowers, reasonably formulate loan amounts and interest rates, thereby reducing default risk and improving the efficiency of credit decisions.
[0003] However, traditional credit reporting relies mainly on offline channels and limited financial institution data sources, resulting in incomplete coverage and difficulty in fully reflecting consumers' credit status. Summary of the Invention
[0004] The purpose of this invention is to provide a consumer finance credit reporting management system to solve the problem that traditional systems mainly rely on offline channels and limited financial institution data sources, resulting in incomplete credit data coverage and difficulty in fully reflecting consumers' credit status.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] A consumer finance credit reporting management system includes:
[0007] The data collection module is used to collect and integrate consumers' multi-source credit information to obtain integrated data;
[0008] The data processing module is used to perform big data processing on the integrated data to obtain processed data;
[0009] A credit scoring module is used to generate a consumer's credit score based on the processed data;
[0010] The tiered management module is used to assess credit risk based on the credit score and classify it, and provide corresponding decision recommendations based on the classification.
[0011] Preferably, the credit information includes: identity information, loan records, repayment records, and overdue records, and the credit information is collected from at least one of banks, credit card companies, government departments, public utilities, lending institutions, e-commerce platforms, and social media platforms.
[0012] Preferably, the data processing module includes: a data cleaning unit, a data integration unit, and a data transformation unit;
[0013] The data cleaning unit is used to detect and correct missing values, outliers, and duplicate data using big data.
[0014] The data integration unit is used to integrate data from different channels and formats using big data to form a unified data view;
[0015] The data transformation unit is used to perform transformation operations on data using big data, including data normalization and data discretization.
[0016] Preferably, generating a consumer's credit score based on the processed data includes:
[0017] Collect consumers' historical credit data and credit score data as a dataset;
[0018] The scoring model is trained using the dataset, and its performance is optimized through feature selection and feature scaling to obtain the trained scoring model.
[0019] The processed data is input into the trained scoring model to generate the consumer's credit score.
[0020] Preferably, generating a consumer's credit score based on the processed data includes:
[0021] The consumer's credit score is generated using the following formula:
[0022] P(Y=1|X)=1 / (1+e ∧ (-(β0+β1X1+β2X2+...+βnXn)))
[0023] Where Y represents the consumer's credit score, X represents the set of the consumer's feature variables, β0 represents the intercept term, (β1, β2, ..., βn) represent the model parameters, and (X1, X2, ..., Xn) represent the consumer's feature variables.
[0024] Preferably, the credit risk includes the consumer's repayment ability, repayment willingness, and fraud risk.
[0025] Preferably, the step of generating a consumer's credit score based on the processed data further includes:
[0026] Select feature variables related to the credit risk prediction from the processed data;
[0027] An attention layer is designed to dynamically assign weights to different feature variables based on the predicted credit risk type, thereby obtaining a weighted feature vector. The weighted feature vector is then input into the scoring model for credit scoring.
[0028] Preferably, the attention layer includes:
[0029] For each predicted credit risk type, a query vector is designed to represent the focus of attention for that risk type;
[0030] The feature variables are converted into key vectors and value vectors. The key vectors are used to calculate the similarity with the query vector, and the value vectors are used to represent the actual contribution of the feature variables.
[0031] The attention weights are calculated using the query vector and the key vector, and these attention weights are used to reflect the importance of different feature variables in predicting this type of risk.
[0032] The value vector is weighted and summed using the attention weights to obtain the weighted feature vector.
[0033] Compared with the prior art, the beneficial effects of the present invention are:
[0034] This invention collects and integrates credit information from multiple consumer data sources to gain a comprehensive understanding of consumers' creditworthiness, providing a data foundation for subsequent analysis. The integrated data then undergoes big data processing to achieve real-time data collection and processing, enabling the timely detection of potential credit risks. A pre-established credit scoring model is then used to score consumers' creditworthiness, predict their future credit performance, and assess their credit risk level. Based on the credit scoring results, corresponding risk control strategies are formulated, facilitating credit management by financial institutions. Attached Figure Description
[0035] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:
[0036] Figure 1 This is a system module block diagram of the present invention. Detailed Implementation
[0037] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0038] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0039] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0040] As attached Figure 1 As shown:
[0041] Example 1: This example provides a consumer finance credit reporting management system, including:
[0042] The data collection module is used to collect and integrate consumers' multi-source credit information to obtain integrated data;
[0043] The data processing module is used to perform big data processing on the integrated data to obtain the processed data;
[0044] The credit scoring module is used to generate a consumer's credit score based on the processed data.
[0045] The tiered management module is used to assess credit risk based on credit scores and classify it into tiers, and then provide corresponding decision-making suggestions based on the classification.
[0046] By collecting and integrating credit information from multiple consumer data sources, a comprehensive understanding of consumers' creditworthiness can be achieved, providing a data foundation for subsequent analysis. The integrated data then undergoes big data processing to enable real-time collection and processing, timely detection of potential credit risks, and standardized data formatting to improve data quality and facilitate subsequent processing and analysis. A well-established credit scoring model is then used to score consumers' creditworthiness, predict their future credit performance, and determine their credit risk level. Based on the credit scoring results, corresponding risk control strategies are formulated, such as setting different credit limits and interest rates or rejecting high-risk consumers, thereby reducing credit risk.
[0047] The policy recommendations include offering more favorable credit terms to consumers with higher credit ratings and implementing stricter risk control measures for consumers with lower credit ratings.
[0048] Specifically, credit information includes: identity information, loan records, repayment records, and overdue records. Credit information is collected from at least one of the following channels: banks, credit card companies, government departments, public utilities, lending institutions, e-commerce platforms, and social media platforms. By collecting data from multiple sources, comprehensive consumer credit information can be obtained, providing a data foundation for subsequent analysis and improving the accuracy of understanding consumers' credit status.
[0049] Specifically, the data processing module includes: a data cleaning unit, a data integration unit, and a data transformation unit;
[0050] The data cleaning unit is used to detect and correct missing values, outliers, and duplicate data using big data.
[0051] The data integration unit is used to integrate data from different channels and formats using big data to form a unified data view;
[0052] The data transformation unit is used to perform transformation operations on big data. The transformation operations include data normalization and data discretization. Data normalization is used to scale data to a specific range, and data discretization is used to convert continuous data into discrete categories, such as dividing age into different age groups.
[0053] Specifically, a consumer's credit score is generated based on the processed data, including:
[0054] Collect consumers' historical credit data and credit score data as a dataset;
[0055] The scoring model is trained using a dataset, and its performance is optimized through feature selection and feature scaling to obtain a well-trained scoring model. Feature selection is used to select feature variables related to credit risk, and feature scaling is used to scale the feature variables, such as standardization or normalization.
[0056] The processed data is then input into the trained rating model to generate a consumer's credit score.
[0057] Specifically, a consumer's credit score is generated based on the processed data, including:
[0058] The consumer's credit score is generated using the following formula:
[0059] P(Y=1|X)=1 / (1+e ∧ (-(β0+β1X1+β2X2+...+βnXn)))
[0060] Where Y represents the consumer's credit score, X represents the set of the consumer's feature variables, β0 represents the intercept term, (β1, β2, ..., βn) represent the parameters of the model, and (X1, X2, ..., Xn) represent the consumer's feature variables. The scoring model can output the consumer's default probability, and the higher the probability, the greater the credit risk.
[0061] Specifically, credit risk includes a consumer's ability to repay, willingness to repay, and fraud risk.
[0062] Example 2: This example provides a consumer finance credit reporting management system, including:
[0063] The data collection module is used to collect and integrate consumers' multi-source credit information to obtain integrated data. The data collection module is the cornerstone of the consumer finance credit management system. It collects consumers' credit information through multiple channels, ensuring the comprehensiveness and diversity of the data. This integration of multi-source data can more accurately depict consumers' credit profiles and provide a solid foundation for subsequent data processing and credit scoring.
[0064] For example, a consumer frequently shops on an e-commerce platform and repays on time, while also having a stable borrowing history at a bank. The data collection module will obtain the consumer's shopping behavior and repayment records from the e-commerce platform, and their borrowing and repayment information from the bank. This information will then be integrated to form the consumer's complete credit information.
[0065] The data processing module is used to perform big data processing on the integrated data to obtain processed data. The data processing module cleans, integrates and transforms the integrated data to ensure the accuracy and consistency of the data.
[0066] For example, if a consumer's loan record is found to have missing values, the data cleaning unit will try to supplement this information through other channels (such as banks or e-commerce platforms), or estimate and fill it based on existing data. In the data integration stage, the loan and repayment records from banks and e-commerce platforms will be integrated to form the consumer's complete loan history. In the data transformation stage, the loan amount will be standardized to make it comparable among different consumers.
[0067] The credit scoring module is used to generate a consumer's credit score based on the processed data. The credit scoring module uses a trained scoring model to score the consumer's credit and predict their future credit performance. The scoring model is trained based on the consumer's historical credit data and credit score data. The scoring results can reflect the consumer's credit risk level and provide a basis for decision-making for the hierarchical management module.
[0068] The tiered management module is used to assess credit risk based on credit scores and classify consumers accordingly. Based on the classification, it provides corresponding decision-making suggestions. The tiered management module assesses the credit risk of consumers based on their credit scores and manages them in a tiered manner. Consumers with higher credit ratings can be given more favorable credit terms, while consumers with lower credit ratings need to be subject to stricter risk control measures. This tiered management strategy can reduce credit risk and improve the robustness of credit business.
[0069] By collecting and integrating credit information from multiple consumer data sources, a comprehensive understanding of consumers' creditworthiness can be achieved, providing a data foundation for subsequent analysis. The integrated data then undergoes big data processing to enable real-time collection and processing, timely detection of potential credit risks, and standardized data formatting to improve data quality and facilitate subsequent processing and analysis. A well-established credit scoring model is then used to score consumers' creditworthiness, predict their future credit performance, and assess their credit risk level. Based on the credit scoring results, corresponding risk control strategies can be developed, such as setting different credit limits and interest rates or rejecting high-risk consumers, thereby reducing credit risk.
[0070] The policy recommendations include offering more favorable credit terms to consumers with higher credit ratings and implementing stricter risk control measures for consumers with lower credit ratings.
[0071] Specifically, credit information includes: identity information, loan records, repayment records, and overdue records. Credit information is collected from at least one of the following channels: banks, credit card companies, government departments, public utilities, lending institutions, e-commerce platforms, and social media platforms. By collecting data from multiple sources, comprehensive consumer credit information can be obtained, providing a data foundation for subsequent analysis and improving the accuracy of understanding consumers' credit status.
[0072] Specifically, the data processing module includes a data cleaning unit, a data integration unit, and a data transformation unit. The data cleaning unit can detect and correct missing values, outliers, and duplicate data to improve data quality. The data integration unit integrates data from different channels and formats into a unified data view to facilitate subsequent analysis. The data transformation unit performs normalization and discretization on the data to make it more suitable for model training.
[0073] The data cleaning unit is used to detect and correct missing values, outliers, and duplicate data using big data.
[0074] The data integration unit is used to integrate data from different channels and formats using big data to form a unified data view;
[0075] The data transformation unit is used to perform transformation operations on big data. The transformation operations include data normalization and data discretization. Data normalization is used to scale data to a specific range, and data discretization is used to convert continuous data into discrete categories, such as dividing age into different age groups.
[0076] Specifically, a consumer's credit score is generated based on the processed data, including:
[0077] Collect consumers' historical credit data and credit score data as a dataset;
[0078] The scoring model is trained using a dataset, and its performance is optimized through feature selection and feature scaling to obtain a well-trained scoring model. Feature selection is used to select feature variables related to credit risk, and feature scaling is used to scale the feature variables, such as standardization or normalization.
[0079] The processed data is then input into the trained rating model to generate a consumer's credit score.
[0080] Specifically, a consumer's credit score is generated based on the processed data, including:
[0081] The consumer's credit score is generated using the following formula:
[0082] P(Y=1|X)=1 / (1+e ∧ (-(β0+β1X1+β2X2+...+βnXn)))
[0083] Where Y represents the consumer's credit score, X represents the set of the consumer's feature variables, β0 represents the intercept term, (β1, β2, ..., βn) represent the parameters of the model, and (X1, X2, ..., Xn) represent the consumer's feature variables. The scoring model can output the consumer's default probability, and the higher the probability, the greater the credit risk.
[0084] Specifically, credit risk includes a consumer's ability to repay, willingness to repay, and fraud risk.
[0085] Specifically, generating consumer credit scores based on processed data also includes:
[0086] Select characteristic variables relevant to credit risk prediction from the processed data;
[0087] An attention layer is designed to dynamically assign weights to different feature variables based on the predicted credit risk type, resulting in a weighted feature vector. This weighted feature vector is then input into the scoring model for credit scoring.
[0088] Specifically, the attention layer includes:
[0089] Design a query vector for each predicted credit risk type to represent the focus of attention for that risk type;
[0090] The feature variables are converted into key vectors and value vectors. The key vectors are used to calculate the similarity with the query vector, and the value vectors are used to represent the actual contribution of the feature variables.
[0091] Attention weights are calculated using query vectors and key vectors. These attention weights reflect the importance of different feature variables in predicting this risk type. The similarity is converted into attention weights using the softmax function.
[0092] The value vector is weighted and summed using attention weights to obtain a weighted feature vector.
[0093] The weighted eigenvectors are calculated using the following formula:
[0094]
[0095] Where Q represents the query vector, used to indicate the focus of the predicted credit risk type, and is a query vector designed for different credit risk types; K represents the key vector; and V represents the value vector. K and V are the key and value vectors transformed from feature variables selected from the processed data. k The dimension of the key vector is used to adjust the scale of similarity calculation. The softmax function is used to convert the similarity function into attention weights, ensuring that the sum of the weights is 1.
[0096] As can be seen from the above, by designing an attention layer, the weights of corresponding feature variables can be dynamically adjusted according to the type of credit risk to be predicted, such as repayment ability, repayment willingness and fraud risk. This makes the prediction model more flexible in responding to different risk prediction scenarios, makes the model pay more attention to the features related to the current risk prediction, and improves the accuracy and pertinence of the prediction results.
[0097] Specifically, for repayment ability risk, the model will pay more attention to characteristics such as income and debt; for repayment willingness risk, the model will pay more attention to characteristics such as historical repayment records and credit history; and for fraud risk, the model will pay more attention to characteristics such as identity verification information and transaction behavior.
[0098] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible (e.g., changes in the size, dimensions, structure, shape, and proportions of various elements, as well as parameter values (e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in this application). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise altered, and the nature or number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of the invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure described herein that performs the function, and not only structurally equivalent but also equivalent in structure. Other substitutions, modifications, alterations, and omissions may be made in the design, operation, and arrangement of the exemplary embodiments without departing from the scope of the invention. Therefore, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.
[0099] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the best mode of carrying out the invention as currently considered, or those features that are not relevant to implementing the invention) may be omitted.
[0100] It should be understood that numerous specific implementation decisions can be made during the development of any practical implementation, such as in any engineering or design project. Such development efforts may be complex and time-consuming, but for those skilled in the art who benefit from this disclosure, the development effort will be a routine work of design, manufacturing, and production without requiring much experimentation.
[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A consumer finance credit reporting management system, characterized in that, include: The data collection module is used to collect and integrate consumers' multi-source credit information to obtain integrated data; The data processing module is used to perform big data processing on the integrated data to obtain processed data; A credit scoring module is used to generate a consumer's credit score based on the processed data; The tiered management module is used to assess credit risk based on the credit score and classify it, and provide corresponding decision recommendations based on the classification.
2. The consumer finance credit reporting management system according to claim 1, characterized in that, The credit information includes: identity information, loan records, repayment records, and overdue records. The channels for collecting the credit information are at least one of banks, credit card companies, government departments, public utilities, lending institutions, e-commerce platforms, and social media platforms.
3. The consumer finance credit reporting management system according to claim 1, characterized in that, The data processing module includes: a data cleaning unit, a data integration unit, and a data transformation unit; The data cleaning unit is used to detect and correct missing values, outliers, and duplicate data using big data. The data integration unit is used to integrate data from different channels and formats using big data to form a unified data view; The data transformation unit is used to perform transformation operations on data using big data, including data normalization and data discretization.
4. The consumer finance credit reporting management system according to claim 1, characterized in that, The process of generating a consumer's credit score based on the processed data includes: Collect consumers' historical credit data and credit score data as a dataset; The scoring model is trained using the dataset, and its performance is optimized through feature selection and feature scaling to obtain the trained scoring model. The processed data is input into the trained scoring model to generate the consumer's credit score.
5. A consumer finance credit reporting management system according to claim 1, characterized in that, The process of generating a consumer's credit score based on the processed data includes: The consumer's credit score is generated using the following formula: P(Y=1|X)=1 / (1+e ∧ (-(β0+β1X1+β2X2+...+βnXn))) Where Y represents the consumer's credit score, X represents the set of the consumer's feature variables, β0 represents the intercept term, (β1, β2, ..., βn) represent the model parameters, and (X1, X2, ..., Xn) represent the consumer's feature variables.
6. A consumer finance credit reporting management system according to claim 1, characterized in that, The credit risk mentioned includes the consumer's ability to repay, willingness to repay, and fraud risk.
7. A consumer finance credit reporting management system according to claim 5, characterized in that, The process of generating a consumer's credit score based on the processed data further includes: Select feature variables related to the credit risk prediction from the processed data; An attention layer is designed to dynamically assign weights to different feature variables based on the predicted credit risk type, thereby obtaining a weighted feature vector. The weighted feature vector is then input into the scoring model for credit scoring.
8. A consumer finance credit reporting management system according to claim 7, characterized in that, The attention layer includes: For each predicted credit risk type, a query vector is designed to represent the focus of attention for that risk type; The feature variables are converted into key vectors and value vectors. The key vectors are used to calculate the similarity with the query vector, and the value vectors are used to represent the actual contribution of the feature variables. The attention weights are calculated using the query vector and the key vector, and these attention weights are used to reflect the importance of different feature variables in predicting this type of risk. The value vector is weighted and summed using the attention weights to obtain the weighted feature vector.