A method, apparatus, equipment and medium for bank user risk analysis
By using the Mann-Shore model and the DeShaw model in combination with the initial credit coefficient to perform weighted calculation and correction on the probability and degree of financial anomalies of bank users, a target risk index is generated. This solves the problem of poor consistency and accuracy of analysis results in existing technologies and achieves more accurate risk analysis.
Patent Information
- Application Number
- CN202611343623.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-01
- Publication Date
- 2026-11-17
AI Technical Summary
The consistency and accuracy of risk analysis results for bank users in existing technologies are poor, mainly because they rely on the professional experience of risk review personnel, which makes the analysis results susceptible to subjective factors.
By acquiring the target user's historical financial data and initial credit score, as well as the corresponding data quality data, the Mann+Dershaw model and the Dershaw model are used for analysis. The initial credit score is then used for correction to generate the target risk index.
It improves the accuracy and consistency of risk analysis for bank users and reduces risk analysis bias caused by missing, incomplete, or unreliable financial data.
Smart Images

Figure CN123118230A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, equipment and medium for risk analysis of bank users. Background Technology
[0002] With the development of fintech, banks and other financial institutions typically need to conduct risk analysis on their users during credit approval, loan management, and risk control processes to assess the creditworthiness and default risk of target users. The results of this risk analysis directly impact credit limit determination, loan approval, and post-loan risk management. Therefore, accurately and efficiently conducting bank user risk analysis has become a crucial research direction in the field of intelligent risk control for banks.
[0003] Currently, existing technologies typically involve risk auditors obtaining historical financial data of target users and combining it with information such as business operations, financial statements, and credit information to conduct a comprehensive analysis of the target user's financial status and operational risks. They then manually determine whether the target user has any financial abnormalities, operational abnormalities, or credit risks, and determine the corresponding risk level of the target user based on the analysis results.
[0004] However, the aforementioned manual analysis methods mainly rely on the professional experience of risk auditors. Different auditors have different understandings of financial indicators, credit information, and abnormal data, which makes the risk analysis results susceptible to subjective factors and results in poor consistency. Summary of the Invention
[0005] This invention provides a method, apparatus, device, and medium for bank user risk analysis. The embodiments of this invention can improve the accuracy of user risk analysis.
[0006] In a first aspect, embodiments of the present invention provide a method for risk analysis of bank users, the method comprising:
[0007] Obtain the target user's historical financial data and initial credit score, as well as the corresponding data quality data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level;
[0008] The historical financial data and data quality data are respectively input into the Mann model and the DeShaw model to obtain the probability of financial anomalies and the degree of financial risk of the target user.
[0009] The initial risk index of the target user is obtained by weighting the probability of financial anomaly and the degree of financial risk.
[0010] Based on the initial credit coefficient, the initial risk index is corrected to obtain the target risk index for the target user.
[0011] Secondly, embodiments of the present invention also provide a bank user risk analysis device, the device comprising:
[0012] The data acquisition module is used to acquire the target user's historical financial data and initial credit score, as well as the data quality data corresponding to the historical financial data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level;
[0013] The data input module is used to input the historical financial data and data quality data into the Mann model and the DeShaw model respectively to obtain the financial anomaly probability and financial risk level of the target user.
[0014] The initial risk index generation module is used to perform a weighted calculation of the financial anomaly probability and the degree of financial risk to obtain the initial risk index of the target user;
[0015] The target risk index generation module is used to correct the initial risk index based on the initial credit coefficient to obtain the target risk index of the target user.
[0016] Thirdly, embodiments of the present invention also provide a bank user risk analysis device, which includes:
[0017] At least one processor; and
[0018] A memory that is communicatively connected to at least one processor; wherein,
[0019] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the bank user risk analysis method according to any embodiment of the present invention.
[0020] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the bank user risk analysis method of any embodiment of the present invention.
[0021] Fifthly, embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the bank user risk analysis method according to any embodiment of the present invention.
[0022] The technical solution of this invention provides a more comprehensive data foundation for subsequent risk calculations by acquiring the target user's historical financial data and initial credit coefficient, as well as the corresponding data quality data. By determining the data quality data based on at least one of data completeness, data missing rate, and data confidence level, the validity of historical financial data can be evaluated, helping to reduce risk analysis bias caused by missing, incomplete, or insufficiently reliable financial data. By inputting the historical financial data and data quality data into the Mann-Stokes model and the Dershaw model respectively, the financial anomalies and financial risks of the target user can be analyzed based on different risk analysis models, obtaining the target user's risk characteristics from different dimensions. By weighting the probability of financial anomalies and the degree of financial risk, the initial risk index of the target user is obtained, which can integrate different risk analysis results and improve the completeness of the risk analysis results. By correcting the initial risk index based on the initial credit coefficient, the risk analysis results can be adjusted in conjunction with the target user's credit foundation, making the obtained target risk index more comprehensively reflect the target user's risk situation. This solves the technical problem of poor consistency and accuracy of analysis results in the prior art, thereby improving the accuracy of user risk analysis.
[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.
[0025] Figure 1 A flowchart of a bank user risk analysis method provided in an embodiment of the present invention;
[0026] Figure 2 A flowchart of a bank user risk analysis method provided in an embodiment of the present invention;
[0027] Figure 3 This is a schematic diagram of the structure of a bank user risk analysis device provided in an embodiment of the present invention;
[0028] Figure 4 This is a schematic diagram of the structure of a bank user risk analysis device provided in an embodiment of the present invention. Detailed Implementation
[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention 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 the invention 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 a 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.
[0031] In the technical solutions of this invention, the acquisition, storage, and application of the target user's historical financial data and initial credit coefficient, as well as the data quality data corresponding to the historical financial data, all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0032] The information collected is information and data authorized by the user or fully authorized by all parties. The collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse.
[0033] Figure 1 This is a flowchart illustrating a bank user risk analysis method provided in an embodiment of the present invention. This embodiment is applicable to bank user risk analysis, and the method can be executed by a bank user risk analysis device, which can be implemented in hardware and / or software.
[0034] See Figure 1 The bank user risk analysis method shown is applied to the server of the task initiator, including:
[0035] S101. Obtain the target user's historical financial data and initial credit score, as well as the data quality data corresponding to the historical financial data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level.
[0036] Historical financial data refers to a set of data reflecting the financial status and operating conditions of a target user over a historical period. This data may include asset data, liability data, revenue data, profit data, cash flow data, and financial indicators calculated from these data. Historical financial data serves as the foundational input in the bank's user risk analysis process, characterizing the target user's financial changes over a past period. Since a target user's financial situation typically changes with their business activities, analyzing historical financial data allows for the identification of trends and abnormal fluctuations in financial indicators. This provides a data basis for subsequently generating target financial data, calculating the probability of financial anomalies, and determining the degree of financial risk.
[0037] The initial credit coefficient can be a numerical parameter representing the target user's basic credit level, reflecting the target user's credit status before risk analysis. The initial credit coefficient is calculated by inputting the target user's historical financial data and data quality data into a pre-set credit prediction model. It can be a standardized numerical parameter used as an adjustment factor in the subsequent risk index revision process.
[0038] Data quality data refers to data used to describe the validity and reliability of historical financial data, reflecting the data quality during collection, storage, and processing. Data quality data includes at least one of data completeness, missing data rate, and data confidence level. Because historical financial data may contain missing data, anomalies, or have insufficient source reliability, introducing data quality data allows for the evaluation of the reliability of historical financial data. Based on the evaluation results, the historical financial data can be corrected to obtain target financial data more suitable for model analysis.
[0039] S102. Input the historical financial data and data quality data into the Mann model and the DeShaw model respectively to obtain the financial anomaly probability and financial risk level of the target user.
[0040] The Mann+4 model can be considered a risk identification model that analyzes user financial anomalies based on financial indicators. It is used to calculate the probability of a user exhibiting abnormal financial behavior based on their financial data. The Mann+4 model extracts at least one target financial indicator from the target financial data and combines multiple target financial indicators according to preset indicator calculation rules to obtain a Mann+4 score for the target user. Furthermore, it determines the probability of financial anomalies for the target user based on the Mann+4 score and a preset probability mapping relationship.
[0041] The Dershaw model can be considered a risk assessment model that analyzes the degree of financial risk of users based on financial characteristics. It is used to identify potential financial risks of target users based on their financial data. The Dershaw model extracts at least one target financial indicator from the target financial data, further extracts features from the target financial indicator to form target risk features, and then calculates a risk score for the target risk features according to preset risk discrimination rules. This Dershaw score is then used as the degree of financial risk corresponding to the target user.
[0042] The financial anomaly probability refers to the probability value representing the likelihood of a target user experiencing financial anomalies. It reflects the degree to which the target user's financial data may exhibit abnormal changes or behaviors. The financial anomaly probability is calculated by the Mann+4 model based on multiple financial indicators corresponding to the target financial data. Specifically, it can be obtained by calculating the Mann+4 score and converting it according to a preset probability mapping relationship.
[0043] The degree of financial risk refers to the evaluation result used to represent the current financial risk level of the target user. It reflects the severity of risks in the target user's business activities, financial status, and changes in financial indicators. The degree of financial risk is obtained by the Dershaw model through feature extraction based on the target financial indicators corresponding to the target financial data, and by calculating a risk score for the generated target risk features based on risk discrimination rules. It can be expressed in the form of a score value or risk level.
[0044] S103. The initial risk index of the target user is obtained by weighting the probability of financial anomaly and the degree of financial risk.
[0045] The initial risk index can be a comprehensive risk assessment parameter obtained by integrating the probability of financial anomalies and the degree of financial risk of the target user. It is used to reflect the preliminary risk level determined by the target user based on the financial anomalies and financial risks.
[0046] S104. Based on the initial credit coefficient, the initial risk index is corrected to obtain the target risk index of the target user.
[0047] The target risk index can be the final risk assessment parameter obtained by combining the initial risk index with the target user's initial credit coefficient, and is used to represent the risk level of the target user after comprehensive analysis.
[0048] As can be seen, in this embodiment, by acquiring the target user's historical financial data and initial credit coefficient, as well as the corresponding data quality data, a more comprehensive data foundation can be provided for subsequent risk calculations. By determining the data quality data based on at least one of data completeness, data missing rate, and data confidence level, the validity of historical financial data can be evaluated, which helps to reduce risk analysis bias caused by missing, incomplete, or insufficiently reliable financial data. By inputting the historical financial data and data quality data into the Mann-Stokes model and the DeShaw model respectively, the target user's financial anomalies and financial risks can be analyzed based on different risk analysis models, obtaining the target user's risk characteristics from different dimensions. By weighting the financial anomaly probability and the financial risk level, the target user's initial risk index is obtained, which can integrate different risk analysis results and improve the completeness of the risk analysis results. By correcting the initial risk index based on the initial credit coefficient, the risk analysis results can be adjusted in conjunction with the target user's credit foundation, so that the obtained target risk index can more comprehensively reflect the target user's risk situation. This solves the technical problem of poor consistency and accuracy of analysis results in the prior art, thereby improving the accuracy of user risk analysis.
[0049] In an optional embodiment, Figure 2 The flowchart of a bank user risk analysis method provided in this embodiment of the invention refines the process of "inputting the historical financial data and data quality data into the Mann+4 model and the Dershaw model respectively to obtain the financial anomaly probability and financial risk level of the target user" into "based on the data quality data, performing credibility correction on the historical financial data to obtain the target financial data corresponding to the historical financial data; inputting the target financial data into the Mann+4 model to obtain the financial anomaly probability of the target user; inputting the target financial data into the Dershaw model to obtain the financial risk level of the target user," thereby improving the operation of bank user risk analysis.
[0050] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments.
[0051] See Figure 2 The bank user risk analysis methods shown include:
[0052] S201. Obtain the target user's historical financial data and initial credit score, as well as the data quality data corresponding to the historical financial data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level.
[0053] S202. Based on the data quality data, the historical financial data is corrected for credibility to obtain the target financial data corresponding to the historical financial data.
[0054] S203. Input the target financial data into the Mann model to obtain the financial anomaly probability of the target user.
[0055] S204. Input the target financial data into the DeShaw model to obtain the financial risk level of the target user.
[0056] S205. The initial risk index of the target user is obtained by weighting the probability of financial anomaly and the degree of financial risk.
[0057] S206. Based on the initial credit coefficient, the initial risk index is corrected to obtain the target risk index of the target user.
[0058] As can be seen, in this embodiment, by correcting the credibility of the historical financial data based on the data quality data, the target financial data corresponding to the historical financial data is obtained. This allows for adjustments to the validity of the original financial data by incorporating information such as the completeness, missing data, and confidence level of the historical financial data, thus reducing the impact of low-quality financial data on subsequent risk analysis. By inputting the target financial data into the Mann-Stokes model, the financial anomaly probability of the target user is obtained. The financial indicators of the target user can be analyzed using the credibility-corrected financial data, improving the reliability of the data used to calculate the financial anomaly probability. Furthermore, by inputting the target financial data into the DeShaw model, the financial risk level of the target user is obtained. Risk characteristics corresponding to the target user can be extracted based on the target financial data, and the overall financial risk of the target user can be evaluated, providing a risk analysis basis for subsequent risk index calculations.
[0059] In some embodiments, inputting the target financial data into a Mann-Rayet model to obtain the probability of financial anomalies of the target user includes:
[0060] Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data;
[0061] According to the index calculation rules corresponding to the Mann model, at least one of the target financial indicators is combined and calculated to obtain the Mann score of the target user.
[0062] Based on a preset probability mapping table, the probability of financial anomalies of the target user is obtained by looking up the Mann score in the probability mapping table.
[0063] Specifically, after acquiring the target financial data, the financial fields in the target financial data can be analyzed according to preset financial indicator extraction rules to determine target financial indicators that reflect the financial status of the target user. After determining at least one target financial indicator, different target financial indicators can be normalized according to the indicator calculation rules corresponding to the Mann+4 model, and each target financial indicator can be calculated according to a preset indicator combination relationship to obtain the Mann+4 score value corresponding to the target user. After obtaining the Mann+4 score value corresponding to the target user, the Mann+4 score value can be matched with the score interval in a preset probability mapping table, and the financial anomaly probability corresponding to the Mann+4 score value can be determined based on the matching result.
[0064] As can be seen, in this embodiment, by extracting at least one target financial indicator corresponding to the target financial data, the target financial data can be converted into a quantitative indicator that reflects the financial status of the target user, providing a data foundation for subsequent financial anomaly analysis based on the Mann+4 model. By combining and calculating at least one of the target financial indicators according to the indicator calculation rules corresponding to the Mann+4 model, the Mann+4 score of the target user can be obtained. This allows for a comprehensive consideration of the correlation between multiple financial indicators, forming a comprehensive score result that can characterize the financial anomaly of the target user. By finding the financial anomaly probability of the target user based on the Mann+4 score using a preset probability mapping table, the Mann+4 score can be converted into a probabilistic risk analysis result, facilitating the quantitative representation of the degree of financial anomaly of the target user.
[0065] In some embodiments, inputting the target financial data into the Dershaw model to obtain the target user's financial risk level includes:
[0066] Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data;
[0067] Feature extraction is performed on at least one of the target financial indicators to obtain the target risk characteristics corresponding to the target user;
[0068] Based on the risk discrimination rules corresponding to the DeShaw model, a risk score is calculated on the target risk feature vector to obtain the DeShaw score value corresponding to the target user;
[0069] The DeShaw score is used as the degree of financial risk corresponding to the target user.
[0070] Target financial data refers to financial data used as input for risk analysis models, obtained by adjusting the reliability of historical financial data based on its corresponding data quality. Target financial data is generated by adjusting historical financial data based on data quality information such as data completeness, missing data rate, and data confidence level.
[0071] The target risk features can refer to a set of data extracted from target financial indicators in the target financial data to characterize the financial risk features of the target user. In this application, the target risk features are generated by the feature extraction process corresponding to the DeShaw model, and may include feature parameters reflecting the target user's profitability, solvency, operational stability, and financial anomalies.
[0072] Risk discrimination rules refer to preset judgment rules used to determine the risk score of a target user based on the target risk characteristics. These rules describe the correspondence between the target risk characteristics and the risk score results. Risk discrimination rules can include calculation methods, scoring methods, and risk level classification methods for different risk characteristics. Through these rules, a comprehensive analysis of the target risk characteristics can be performed to obtain the corresponding DeShaw score.
[0073] Risk scoring calculation refers to the process of quantifying the risk profile of a target user based on target risk characteristics and risk discrimination rules, generating a score that represents the risk level. The risk scoring calculation is performed by the Dershaw model, which comprehensively processes multiple risk factors in the target risk characteristics to obtain the Dershaw score corresponding to the target user.
[0074] The Dershaw score, in particular, is a numerical result obtained by calculating a risk score for the target user's corresponding target risk characteristics using the Dershaw model. It is used to represent the target user's financial risk level. The Dershaw score is calculated based on the risk discrimination rules corresponding to the Dershaw model after feature extraction of target financial indicators from the target financial data to form target risk characteristics. It can serve as a quantitative indicator to measure the degree of financial risk of the target user.
[0075] As can be seen, in this embodiment, by extracting at least one target financial indicator corresponding to the target financial data, the target financial data can be converted into indicator information that reflects the financial status of the target user, providing a data foundation for subsequent financial risk characteristic analysis; by extracting features from at least one of the target financial indicators, the target risk characteristics corresponding to the target user can be obtained, and information related to financial risk can be extracted from the target financial indicators, enabling the target financial data to be represented in a feature form suitable for risk analysis; by calculating the risk score of the target risk feature vector according to the risk discrimination rule corresponding to the Dershaw model, the Dershaw score value corresponding to the target user can be obtained, and the target risk characteristics can be comprehensively evaluated based on the preset risk judgment logic to form a scoring result that can quantitatively represent the financial risk situation of the target user; by using the Dershaw score value as the degree of financial risk corresponding to the target user, the scoring result calculated by the model can be converted into the financial risk evaluation parameter of the target user, providing a risk analysis basis for subsequent risk index calculation.
[0076] In some embodiments, a weighted calculation is performed on the probability of financial anomaly and the degree of financial risk to obtain an initial risk index for the target user, including:
[0077] Obtain the first weight coefficient and the second weight coefficient; the first weight coefficient is the weight corresponding to the probability of financial anomaly, and the second weight coefficient is the weight corresponding to the degree of financial risk;
[0078] Based on the first weighting coefficient and the second weighting coefficient, the probability of financial anomaly and the degree of financial risk are weighted and calculated to obtain the initial risk index of the target user.
[0079] The first weighting coefficient can be a parameter used to represent the proportion of influence of the probability of financial anomalies in the initial risk index calculation process, and is used to adjust the degree of contribution of the probability of financial anomalies to the final risk analysis result.
[0080] The second weighting coefficient can be a parameter used to represent the proportion of financial risk in the initial risk index calculation process, and is used to adjust the degree of influence of financial risk on the overall risk result.
[0081] As can be seen, in this embodiment, by obtaining the first weighting coefficient and the second weighting coefficient, the influence ratio of the probability of financial anomaly and the degree of financial risk in the initial risk index calculation process can be determined, providing a parameter basis for subsequent risk result fusion; by setting the first weighting coefficient to correspond to the probability of financial anomaly and the second weighting coefficient to correspond to the degree of financial risk, different risk factors can be differentiated according to their importance, so that risk analysis results from different sources can participate in the comprehensive calculation according to a preset ratio; by weighting the probability of financial anomaly and the degree of financial risk according to the first weighting coefficient and the second weighting coefficient, the initial risk index of the target user can be obtained, and risk analysis results from different dimensions can be fused to form an evaluation index for representing the comprehensive risk level of the target user.
[0082] In some embodiments, the initial credit score is obtained in the following manner:
[0083] Obtain the target user's historical financial data;
[0084] The historical financial data and data quality data are input into a preset credit prediction model to obtain the initial credit coefficient corresponding to the target user.
[0085] A credit prediction model can refer to a model used to predict and analyze the creditworthiness of a target user based on their historical financial data and data quality data. It generates an initial credit coefficient for the target user. A credit prediction model can include a data input layer, a feature analysis process, and a credit result output process. Based on the input historical financial data and corresponding data quality data, it extracts feature information that reflects the target user's credit status and generates an initial credit coefficient based on preset prediction rules.
[0086] As can be seen, in this embodiment, by acquiring the historical financial data of the target user, basic data for analyzing the target user's credit level can be obtained, providing a data foundation for subsequent credit prediction model credit feature analysis; by inputting the historical financial data and data quality data into the preset credit prediction model, the credit status of the target user can be comprehensively analyzed by considering historical financial information and corresponding data quality conditions, reducing the impact of differences in financial data quality on the credit prediction results; by obtaining the initial credit coefficient corresponding to the target user based on the credit prediction model, the target user's credit level can be converted into a quantifiable parameter, providing a basis for subsequent correction of the initial risk index using the initial credit coefficient.
[0087] Figure 3This invention provides a schematic diagram of a bank user risk analysis device. This invention is applicable to bank user risk analysis and can execute bank user risk analysis methods. The device can be implemented in hardware and / or software.
[0088] See Figure 3 The bank user risk analysis device shown includes: a data acquisition module 301, a data input module 302, an initial risk index generation module 303, and a target risk index generation module 304, wherein...
[0089] The data acquisition module 301 is used to acquire the target user's historical financial data and initial credit score, as well as the data quality data corresponding to the historical financial data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level;
[0090] Data input module 302 is used to input the historical financial data and data quality data into the Mann model and the DeShaw model respectively to obtain the financial anomaly probability and financial risk level of the target user;
[0091] The initial risk index generation module 303 is used to perform a weighted calculation of the financial anomaly probability and the degree of financial risk to obtain the initial risk index of the target user.
[0092] The target risk index generation module 304 is used to correct the initial risk index based on the initial credit coefficient to obtain the target risk index of the target user.
[0093] The technical solution of this invention provides a more comprehensive data foundation for subsequent risk calculations by acquiring the target user's historical financial data and initial credit coefficient, as well as the corresponding data quality data. By determining the data quality data based on at least one of data completeness, data missing rate, and data confidence level, the validity of historical financial data can be evaluated, helping to reduce risk analysis bias caused by missing, incomplete, or insufficiently reliable financial data. By inputting the historical financial data and data quality data into the Mann-Stokes model and the Dershaw model respectively, the financial anomalies and financial risks of the target user can be analyzed based on different risk analysis models, obtaining the target user's risk characteristics from different dimensions. By weighting the probability of financial anomalies and the degree of financial risk, the initial risk index of the target user is obtained, which can integrate different risk analysis results and improve the completeness of the risk analysis results. By correcting the initial risk index based on the initial credit coefficient, the risk analysis results can be adjusted in conjunction with the target user's credit foundation, making the obtained target risk index more comprehensively reflect the target user's risk situation. This solves the technical problem of poor consistency and accuracy of analysis results in the prior art, thereby improving the accuracy of user risk analysis.
[0094] In some embodiments, in inputting the historical financial data and data quality data into the Mann-Stokes model and the DeShaw model, respectively, to obtain the financial anomaly probability and financial risk level of the target user, the data input module 302 is specifically used for:
[0095] Based on the data quality data, the historical financial data is reconciled to obtain the target financial data corresponding to the historical financial data.
[0096] The target financial data is input into the Mann model to obtain the probability of financial anomalies of the target user;
[0097] The target financial data is input into the DeShaw model to obtain the financial risk level of the target user.
[0098] In some embodiments, in inputting the target financial data into the Mann model to obtain the probability of financial anomalies of the target user, the data input module 302 is specifically used for:
[0099] Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data;
[0100] According to the index calculation rules corresponding to the Mann model, at least one of the target financial indicators is combined and calculated to obtain the Mann score of the target user.
[0101] Based on a preset probability mapping table, the probability of financial anomalies of the target user is obtained by looking up the Mann score in the probability mapping table.
[0102] In some embodiments, in inputting the target financial data into the DeShaw model to obtain the financial risk level of the target user, the data input module 302 is specifically used for:
[0103] Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data;
[0104] Feature extraction is performed on at least one of the target financial indicators to obtain the target risk characteristics corresponding to the target user;
[0105] Based on the risk discrimination rules corresponding to the DeShaw model, a risk score is calculated on the target risk feature vector to obtain the DeShaw score value corresponding to the target user;
[0106] The DeShaw score is used as the degree of financial risk corresponding to the target user.
[0107] In some embodiments, in obtaining the initial risk index of the target user by weighting the probability of financial anomaly and the degree of financial risk, the initial risk index generation module 303 is specifically used for:
[0108] Obtain the first weight coefficient and the second weight coefficient; the first weight coefficient is the weight corresponding to the probability of financial anomaly, and the second weight coefficient is the weight corresponding to the degree of financial risk;
[0109] Based on the first weighting coefficient and the second weighting coefficient, the probability of financial anomaly and the degree of financial risk are weighted and calculated to obtain the initial risk index of the target user.
[0110] In some embodiments, the initial credit score is obtained in the following way:
[0111] Obtain the target user's historical financial data;
[0112] The historical financial data and data quality data are input into a preset credit prediction model to obtain the initial credit coefficient corresponding to the target user.
[0113] The bank user risk analysis device provided in this embodiment of the invention can execute the bank user risk analysis method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the bank user risk analysis method.
[0114] Figure 4 This is a schematic diagram of the structure of a bank user risk analysis device provided in an embodiment of the present invention.
[0115] like Figure 4 As shown, the bank user risk analysis device 400 includes at least one processor 401 and a memory, such as a read-only memory (ROM) 402 and a random access memory (RAM) 403, communicatively connected to the at least one processor 401. The memory stores computer programs executable by the at least one processor. The processor 401 can perform various appropriate actions and processes based on the computer program stored in the ROM 402 or loaded into the RAM 403 from the storage unit 408. The RAM 403 can also store various programs and data required for the operation of the bank user risk analysis device 400. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.
[0116] Multiple components in the bank user risk analysis device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, optical disk, etc.; and a communication unit 409, such as a network card, modem, wireless transceiver, etc. The communication unit 409 allows the bank user risk analysis device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0117] Processor 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 401 performs the various methods and processes described above, such as bank user risk analysis methods.
[0118] In some embodiments, the bank user risk analysis method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the bank user risk analysis device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by processor 401, one or more steps of the bank user risk analysis method described above may be performed. Alternatively, in other embodiments, processor 401 may be configured to execute the bank user risk analysis method by any other suitable means (e.g., by means of firmware).
[0119] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0120] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0121] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0122] To provide user interaction, the systems and techniques described herein can be implemented on a bank user risk analysis device, which includes: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the bank user risk analysis device. Other types of devices can also be used to provide user interaction; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, speech input, or tactile input).
[0123] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0124] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.
[0125] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.
[0126] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A method for analyzing bank user risk, characterized in that, The method, applied to a server of the task initiator, includes: Obtain the target user's historical financial data and initial credit score, as well as the corresponding data quality data; the data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level; The historical financial data and data quality data are respectively input into the Mann model and the DeShaw model to obtain the probability of financial anomalies and the degree of financial risk of the target user. The initial risk index of the target user is obtained by weighting the probability of financial anomaly and the degree of financial risk. Based on the initial credit coefficient, the initial risk index is corrected to obtain the target risk index for the target user.
2. The method according to claim 1, characterized in that, The step of inputting the historical financial data and data quality data into the Mann-Stokes model and the Dershaw model, respectively, to obtain the financial anomaly probability and financial risk level of the target user includes: Based on the data quality data, the historical financial data is reconciled to obtain the target financial data corresponding to the historical financial data. The target financial data is input into the Mann model to obtain the probability of financial anomalies of the target user; The target financial data is input into the DeShaw model to obtain the financial risk level of the target user.
3. The method according to claim 2, characterized in that, The step of inputting the target financial data into the Mann-Stokes model to obtain the probability of financial anomalies of the target user includes: Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data; According to the index calculation rules corresponding to the Mann model, at least one of the target financial indicators is combined and calculated to obtain the Mann score of the target user. Based on a preset probability mapping table, the probability of financial anomalies of the target user is obtained by looking up the Mann score in the probability mapping table.
4. The method according to claim 2, characterized in that, The step of inputting the target financial data into the Dershaw model to obtain the target user's financial risk level includes: Based on the target financial data, extract at least one target financial indicator corresponding to the target financial data; Feature extraction is performed on at least one of the target financial indicators to obtain the target risk characteristics corresponding to the target user; Based on the risk discrimination rules corresponding to the DeShaw model, a risk score is calculated on the target risk feature vector to obtain the DeShaw score value corresponding to the target user; The DeShaw score is used as the degree of financial risk corresponding to the target user.
5. The method according to claim 1, characterized in that, The weighted calculation of the financial anomaly probability and the degree of financial risk to obtain the initial risk index of the target user includes: Obtain the first weight coefficient and the second weight coefficient; the first weight coefficient is the weight corresponding to the probability of financial anomaly, and the second weight coefficient is the weight corresponding to the degree of financial risk; Based on the first weighting coefficient and the second weighting coefficient, the probability of financial anomaly and the degree of financial risk are weighted and calculated to obtain the initial risk index of the target user.
6. The method according to claim 1, characterized in that, The initial credit score is obtained in the following way: Obtain the target user's historical financial data; The historical financial data and data quality data are input into a preset credit prediction model to obtain the initial credit coefficient corresponding to the target user.
7. A bank user risk analysis device, characterized in that, include: The data acquisition module is used to acquire the target user's historical financial data and initial credit score, as well as the data quality data corresponding to the historical financial data; The data quality data includes at least one of the following: data completeness, data missing rate, and data confidence level; The data input module is used to input the historical financial data and data quality data into the Mann model and the DeShaw model respectively to obtain the financial anomaly probability and financial risk level of the target user. The initial risk index generation module is used to perform a weighted calculation of the financial anomaly probability and the degree of financial risk to obtain the initial risk index of the target user; The target risk index generation module is used to correct the initial risk index based on the initial credit coefficient to obtain the target risk index of the target user.
8. A bank user risk analysis device, characterized in that, The bank user risk analysis device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the bank user risk analysis method according to any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the bank user risk analysis method according to any one of claims 1-6.
10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the bank user risk analysis method according to any one of claims 1-6.
