Financing risk monitoring method and device based on credit asset batch classification

Abstract

The present invention provides a financing risk monitoring method and device based on batch classification of credit assets, which relates to the field of financial technology. The method comprises: performing batch initial classification of debt items on basic data and debt item data of users; performing batch cross-adjustment on the debt item data after batch initial classification to obtain classified debt item data; determining whether the classified debt item data has abnormal risks, and if so, feeding back abnormal information to management personnel. The present invention can perform batch classification of user credit assets, and financing risk monitoring improves the classification efficiency, asset security and risk monitoring capabilities of credit assets.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, in particular to the field of financial technology, and more particularly to a financing risk monitoring method and device based on batch classification of credit assets. Background Art

[0002] Credit asset classification includes asset quality classification and risk classification, which refers to the process of classifying corporate clients' on-balance sheet assets into twelve levels and off-balance sheet credit assets into five levels based on the judgment of risk level and expected credit loss of specific debts, and classifying risk assets at the same time, so as to select corresponding financial instrument measurement methods for risk provision management according to corresponding accounting standards. The essence of credit asset classification is to judge the possibility of debtors repaying the principal and interest of credit assets in a timely and full manner. The main purpose is to reveal the actual value and risk level of credit assets, and reflect their quality in a real, comprehensive and dynamic manner. Credit asset classification is the daily work of financial institutions such as banks. The handlers need to track and update the production, operation and financial information of borrowers in a timely manner, and monitor and adjust the changes in credit asset risks in real time according to changes in the debtor's operation and management.

[0003] With the continuous changes in the internal and external environment of the financial industry and the rapid development of financing products, customers have many financing needs, diverse scenarios, and strong timeliness requirements. The on-balance sheet and off-balance sheet financing products that various financial institutions have innovated, developed, and vigorously promoted are also gradually moving towards online, automated, and batch production. At present, the traditional method of classifying credit assets according to debt items, although some information is collected automatically, but due to the large number of debt items, the handling staff need to spend a lot of time on repetitive work, the labor cost of classification operations is high, and the user experience is not good. At the same time, manual operations with large business volumes may lead to inconsistent classification results for the same customer or contract due to operational errors, and the convenience of classification operations is low. Summary of the invention

[0004] One object of the present invention is to provide a financing risk monitoring method based on batch classification of credit assets, to classify user credit assets in batches, and to improve the classification efficiency, asset security and risk monitoring capabilities of credit assets. Another object of the present invention is to provide a financing risk monitoring device based on batch classification of credit assets. Another object of the present invention is to provide a computer device. Another object of the present invention is to provide a readable medium.

[0005] In order to achieve the above objectives, the present invention discloses, on one hand, a financing risk monitoring method based on batch classification of credit assets, comprising:

[0006] Carry out initial batch classification of debts based on the user's basic data and debt data;

[0007] Perform batch cross-adjustment on the debt item data after batch initial classification to obtain classified debt item data;

[0008] Determine whether the classified debt data has abnormal risks. If so, feedback the abnormal information to the management personnel.

[0009] Preferably, the method further comprises, before performing initial batch classification of the basic data and debt item data of the user:

[0010] Obtain customer information and debt information stored by customers in the trading system;

[0011] Obtain external data information from external systems through interfaces;

[0012] The customer information, debt information and external data information are decomposed and packaged to obtain the basic data and debt data of the user.

[0013] Preferably, the method further comprises, before performing initial batch classification of the basic data and debt item data of the user:

[0014] The basic data and debt data of the corresponding user are determined according to the screening dimensions input by the staff, and the screening dimensions include at least one of the time dimension, the branch dimension and the user dimension.

[0015] Preferably, the method further comprises, before performing initial batch classification of the basic data and debt item data of the user:

[0016] Receive debt modification requests from staff;

[0017] Modify the debt information of the corresponding target debt in the debt data according to the debt modification request;

[0018] The debt information of other debts of the user to which the target debt belongs is modified according to the debt modification request.

[0019] Preferably, the initial batch classification of the basic data and debt data of the user specifically includes:

[0020] The initial classification model is used to conduct initial classification of debts in batches based on the user's basic data and debt data and the customer evaluation index system and the debt evaluation index system.

[0021] Preferably, the batch cross-adjustment of the debt item data after the batch initial classification to obtain the classified debt item data specifically includes:

[0022] If all debt items after the initial batch classification are classified as non-performing or impaired, the batch classification is performed according to the single contract, and the classification result obtained is the worst classification result of all debt items under the contract;

[0023] If the debt data after the initial batch classification is not all non-performing or impaired, then determine whether the customer's debts are completely non-performing or impaired. If so, then batch classification is performed according to the single contract to ensure that the debt classification results under the same contract are consistent and are the worst classification results; if not, then all the customer's debts are batch classified according to the single customer to ensure that the debt classification results for the same customer are consistent and are the worst classification results among all the customer's debts.

[0024] Preferably, the determining whether the classified debt data has abnormal risk specifically includes:

[0025] Conduct transaction risk detection on classified debt data of classification results through scoring model;

[0026] If abnormal transaction risks are detected, the abnormal information will be fed back to the management staff.

[0027] The present invention also discloses a financing risk monitoring device based on credit asset batch classification, comprising:

[0028] The debt initial classification module is used to perform initial classification of debts in batches based on the user's basic data and debt data;

[0029] The cross adjustment module is used to perform batch cross adjustment on the debt item data after batch initial classification to obtain classified debt item data;

[0030] The abnormality monitoring module is used to determine whether the classified debt data has abnormal risks. If so, the abnormal information is fed back to the management personnel.

[0031] The present invention also discloses a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor.

[0032] When the processor executes the program, the method described above is implemented.

[0033] The present invention also discloses a computer readable medium having a computer program stored thereon.

[0034] When the program is executed by a processor, the method described above is implemented.

[0035] The financing risk monitoring method based on the batch classification of credit assets of the present invention performs batch initial classification of debts on the user's basic data and debt data, performs batch cross-adjustment on the debt data after batch initial classification to obtain classified debt data, and determines whether the classified debt data has abnormal risks. If so, abnormal information is fed back to the management personnel. It can be seen that the present invention automatically performs batch initial classification and batch cross-adjustment of debts, which can effectively avoid the problem of inconsistent classification results under the same customer or contract due to subjective or objective operational reasons of staff when classifying by single assets in the traditional way, reduce operational risks, and at the same time, financing risk monitoring improves the classification efficiency, asset security and risk monitoring capabilities of credit assets. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A flowchart showing a specific embodiment of the financing risk monitoring method based on credit asset batch classification of the present invention;

[0038] Figure 2 A flowchart showing a specific embodiment of the financing risk monitoring method based on batch classification of credit assets of the present invention for obtaining basic data and debt item data;

[0039] Figure 3 A flow chart showing the modification of debt item data according to a specific embodiment of the financing risk monitoring method based on batch classification of credit assets of the present invention;

[0040] Figure 4 A flowchart showing a specific embodiment S200 of the financing risk monitoring method based on batch classification of credit assets of the present invention;

[0041] Figure 5 A flowchart showing a specific embodiment S300 of the financing risk monitoring method based on batch classification of credit assets of the present invention;

[0042] Figure 6 A structural diagram showing a specific embodiment of a financing risk monitoring device based on credit asset batch classification according to the present invention;

[0043] Figure 7 A structural diagram showing a specific embodiment of the financing risk monitoring device based on batch classification of credit assets of the present invention includes a data processing module;

[0044] Figure 8A schematic diagram showing the structure of a computer device suitable for implementing an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be noted that the financing risk monitoring method and device based on batch classification of credit assets disclosed in the present application can be used in the field of financial technology, and can also be used in any field outside the field of financial technology. The application field of the financing risk monitoring method and device based on batch classification of credit assets disclosed in the present application is not limited.

[0047] In order to facilitate the understanding of the technical solution provided by this application, the relevant contents of the technical solution of this application are first described below. The financing risk monitoring method based on batch classification of credit assets provided by the embodiment of the present invention can automatically perform batch initial classification and batch cross-adjustment of debt items, effectively avoiding the problem of inconsistent classification results under the same customer or contract due to subjective or objective operational reasons of staff when classifying by single asset in the traditional way, reducing operational risks, and at the same time, financing risk monitoring improves the classification efficiency, asset security and risk monitoring capabilities of credit assets.

[0048] The following takes the financing risk monitoring device based on batch classification of credit assets as an example of the execution subject to illustrate the implementation process of the financing risk monitoring method based on batch classification of credit assets provided by the embodiment of the present invention. It can be understood that the execution subject of the financing risk monitoring method based on batch classification of credit assets provided by the embodiment of the present invention includes but is not limited to the financing risk monitoring device based on batch classification of credit assets.

[0049] According to one aspect of the present invention, this embodiment discloses a financing risk monitoring method based on credit asset batch classification. Figure 1 As shown, in this embodiment, the method includes:

[0050] S100: Perform initial batch classification of the user's basic data and debt data.

[0051] S200: Perform batch cross-adjustment on the debt item data after initial batch classification to obtain classified debt item data.

[0052] S300: Determine whether the classified debt data has abnormal risks, and if so, feedback the abnormal information to the management personnel.

[0053] The financing risk monitoring method based on the batch classification of credit assets of the present invention performs batch initial classification of debts on the user's basic data and debt data, performs batch cross-adjustment on the debt data after batch initial classification to obtain classified debt data, and determines whether the classified debt data has abnormal risks. If so, abnormal information is fed back to the management personnel. It can be seen that the present invention automatically performs batch initial classification and batch cross-adjustment of debts, which can effectively avoid the problem of inconsistent classification results under the same customer or contract due to subjective or objective operational reasons of staff when classifying by single assets in the traditional way, reduce operational risks, and at the same time, financing risk monitoring improves the classification efficiency, asset security and risk monitoring capabilities of credit assets.

[0054] In a preferred embodiment, Figure 2 As shown, the method further includes, before performing initial batch classification of debts on the user's basic data and debt data:

[0055] S011: Obtain the customer information and debt information stored by the customer in the trading system.

[0056] S012: Obtain external data information from the external system through the interface.

[0057] S013: Decomposing and packaging the customer information, debt information and external data information to obtain the basic data and debt data of the user.

[0058] Specifically, when a customer conducts asset transactions through the trading system, the trading system stores the customer's customer information, debt information and other data information. Furthermore, in order to ensure the comprehensiveness of user data, external data information can be obtained from the external system through the interface. The external data information may include the total amount of external guarantees, other banks' quality classification information and other external data information. The data from different data sources are decomposed, packaged, classified and integrated to obtain the user's basic data and debt data, so as to provide it to other processes such as credit asset initial classification for call and calculation.

[0059] In a preferred embodiment, the method further comprises, before performing initial batch classification of the basic data and debt item data of the user:

[0060] S021: Determine the basic data and debt data of the corresponding user according to the screening dimensions input by the staff, wherein the screening dimensions include at least one of the time dimension, the branch dimension and the user dimension.

[0061] Specifically, the relevant data required for asset classification can be loaded to provide data preparation for the subsequent initial asset classification. At the same time, the loaded data information will be provided and displayed to the staff as the basic situation. The loading method may include manual single customer data loading and batch loading of bank data. Among them, the user's basic data may include basic customer information such as credit rating, industry, credit usage and default probability, and debt data may include basic debt information such as business balance, remaining debt term, restructuring situation, issuance date, basic transaction background and overdue (arrears) situation, as well as repayment data information such as second repayment source information such as mortgage guarantee data and guarantor information, to provide data preparation for the subsequent initial asset classification.

[0062] Furthermore, in order to facilitate the monitoring, modification and adjustment of the staff, the asset classification initial classification page can be displayed to the staff, and the customers to be classified under the institution can be queried and displayed with time and branch as the dimensions. The query results will display the business classification overview of each customer of the designated institution in the designated clearing month in the form of a list, and the single-household information can also be directly queried with the customer as the dimension. The staff selects the corresponding customer in the list of customers to be classified to automatically obtain the loaded customer basic data and debt data and other data information.

[0063] In a preferred embodiment, Figure 3 As shown, it further includes before the initial batch classification of the user's basic data and debt data:

[0064] S031: Receive debt modification requests from staff.

[0065] S032: Modify the debt information of the corresponding target debt in the debt data according to the debt modification request.

[0066] S033: Modify the debt information of other debts of the user to which the target debt belongs according to the debt modification request.

[0067] Specifically, if the data information changes, the classified debt data can be manually modified to provide more accurate data preparation for subsequent initial classification. If a single customer involves multiple debts, when modifying the information of one debt, the information of other debts can be updated in batches to achieve adjustments in the customer dimension. For example, when the debt information of the guarantor information in multiple debts in the debt data is associated and unified, when the staff modifies the guarantor information of one debt, the guarantor information of other debts associated with the debt will be automatically updated.

[0068] In a preferred embodiment, the step S100 of performing initial batch classification of the basic data and debt item data of the user may specifically include:

[0069] S110: Performing a batch preliminary classification of debts on the user's basic data and debt data based on the customer evaluation index system and the debt evaluation index system through the preliminary classification model.

[0070] Specifically, the initial classification model can be used to measure the customer's default risk and the specific transaction risk of the debt. The two-step classification model of "customer classification + debt adjustment" is mainly adopted to perform qualitative and quantitative scoring on the classified debts, so as to obtain the corresponding twelve-level classification and three-stage classification results. Among them, the customer evaluation index system in the initial classification model mainly automatically evaluates the customer's behavior characteristics and calculates the customer classification result score through the processes of loss loan prediction, low-risk business prediction, loan commitment part initial classification, off-balance sheet business prediction, country risk adjustment, customer scale and economic composition adjustment, repayment willingness adjustment, customer status adjustment, credit line usage adjustment, etc. The debt evaluation index system adjustment is mainly based on the customer classification result score, through the processes of second repayment source adjustment, remaining term adjustment, actual loan use adjustment, extension adjustment and refinancing and restructuring adjustment, etc., to automatically extract debt characteristics in batches, and use the intersection of each process constraint as the final constraint score of the debt. If the intersection is empty, the upper limit of the set is taken as the final constraint.

[0071] In a preferred embodiment, Figure 4 As shown, the step S200 performs batch cross adjustment on the debt item data after batch initial classification to obtain classified debt item data, specifically including:

[0072] S210: If all the debt data after the initial batch classification are non-performing or impaired, batch classification is performed according to a single contract, and the classification result obtained is the worst classification result among all the debt data under the contract.

[0073] S220: If the debt data after the initial batch classification are not all bad or impaired, determine whether the customer's debts are completely non-bad or impaired. If so, perform batch classification according to the single contract to ensure that the classification results of debts under the same contract are consistent and are the worst classification results; if not, all the customer's debts are batch classified according to the single customer to ensure that the classification results of debts under the same customer are consistent and are the worst classification results among all the customer's debts.

[0074] Specifically, considering the mutual influence between different debts under the same customer, the system's initial classification results are further corrected to form a batch cross-adjustment process for customers. If all the debts that the customer has completed initial classification are non-performing or all are impaired, batch classification is performed according to a single contract, and the classification results are adjusted to the worst classification results (including asset quality classification and risk classification) among all debts under the contract to ensure that the classification results of debts under the same contract are consistent. If the customer has completed initial classification of debts, and the initial classification results are not all non-performing or all are impaired, then continue to determine whether the customer's debts are completely non-performing or impaired. If the customer has completed initial classification of debts, and the initial classification results do not completely contain non-performing or impaired debts, the low-risk debts among them will not participate in the customer rule adjustment. It is only necessary to perform batch classification according to a single contract to ensure that the classification results of debts under the same contract are consistent and the worst classification results are obtained. The asset classification level of low-risk debts is allowed to be inconsistent with the classification level of debts under other contracts of the customer. If the customer has completed the initial classification of debts, and the initial classification results include non-performing or impaired debts, but not all of them are non-performing or impaired debts, all debts of the customer will be classified in batches according to the single customer to ensure that the classification results of debts under the same customer are consistent and the worst classification results among all debts of the customer. After the above judgment process, considering the mutual influence between customer credit assets, the worst adjustment score and grade among all adjustment results are output as the cross-adjustment result, and the system batch cross-adjustment is completed.

[0075] In order to ensure the accuracy of the batch cross-adjustment results, the classification results of the batch cross-adjustment can be fed back to the staff so that the staff can confirm the classification approval results and review the authenticity, validity, accuracy and compliance of the classification results. If there is any objection to the results, or other emergencies occur after the initial classification and the classification results must be adjusted, relevant adjustments can be made to reduce operational risks. Specifically, the handling staff can judge whether to agree with the system classification results based on their own understanding of the actual situation of the customer and the debt, combined with the results obtained by the batch classification and cross-adjustment of the system model. If the system classification results are agreed, the results can be identified and the subsequent process can be carried out. If the handling staff does not recognize the system classification results after the cross-adjustment, the system classification results can be adjusted manually and identified under the premise that the conditions are met. The handling staff will submit the identified system classification results for approval in batches. The reviewer will conduct a batch review of the authenticity, validity, accuracy and compliance of the customer's asset classification results. If there is any objection to the results identified in the previous process, or other emergencies occur after the initial classification and the classification results must be adjusted, the classification results can be adjusted again in batches during the review process. The reviewer conducts batch review of the client's asset classification results. During the review process, the classification results can be adjusted again in batches to complete the approval process.

[0076] In a preferred embodiment, Figure 5As shown, the step S300 of determining whether the classified debt data has abnormal risk specifically includes:

[0077] S310: Perform transaction risk detection on the classified debt data of the classification results through a scoring model.

[0078] S320: If it is detected that there is a risk of abnormal transaction, the abnormal information is fed back to the management personnel.

[0079] Specifically, by using the scoring model to detect the transaction risk of the classified debt data of the classification results, the user's default risk and the transaction risk of the debt can be measured. If there is an abnormal transaction risk, the abnormal information can be fed back to the management personnel, and the abnormal debt can be transferred for disposal and management to monitor the financing risk of the user's credit assets. Among them, the scoring model can be trained by machine learning or neural network and other technologies based on historical normal debt data, which will not be repeated here.

[0080] The present invention is further described below by a specific example. In this specific example, the financing risk monitoring method based on credit asset batch classification includes the following steps:

[0081] S101: Obtain the user's basic data and debt data. The basic data and debt data can be loaded by the user in batches to load the data required for asset classification, or can be loaded by the handling personnel on a customer-by-customer basis. The user's basic data includes basic customer information such as credit rating, industry, credit usage and default probability, and debt data includes basic debt information such as business balance, remaining debt term, restructuring status, issuance date, basic transaction background and overdue (arrears) status, as well as repayment data information such as second repayment source information such as mortgage guarantee data and guarantor information, to provide data preparation for subsequent initial asset classification.

[0082] S102: Display the asset classification preliminary classification page to the staff, query and display the customers to be classified under the institution based on time and branch, and the query result displays the business classification overview of each customer of the designated institution in the designated clearing month in the form of a list (such as the total number of businesses handled, the total number of businesses loaded, the total number of classified businesses, etc.). You can also query single-account information directly based on the customer dimension.

[0083] S103: The staff selects the corresponding customer from the list of customers to be classified, and automatically obtains the loaded customer basic data, debt data and other data information.

[0084] S104: If the data information changes, the loaded data information can be manually modified through the modification and maintenance function to provide more accurate data preparation for the subsequent initial division. If a single customer involves multiple debts, when modifying the information of one debt, the information of other debts can be updated in batches to achieve the adjustment of customer dimensions. If the data information at the time of initial division has not changed compared to that at the time of loading, this step can be skipped.

[0085] S105: The staff clicks the "Initial Classification" button to call the initial classification model, and makes preliminary adjustments to the user's basic data and debt data based on the customer evaluation index system and the debt evaluation index system through the initial classification model, triggering the system batch initial classification of the debts that have completed data preparation, including customer evaluation and debt preliminary adjustment. The customer evaluation index system in the initial classification model mainly automatically evaluates customer behavior characteristics and calculates customer classification result scores through processes such as loss loan prediction, low-risk business prediction, loan commitment part initial classification, off-balance sheet business prediction, country risk adjustment, customer scale and economic composition adjustment, repayment willingness adjustment, customer status adjustment, and credit line usage adjustment. The debt evaluation index system adjustment is mainly based on the customer classification result score, through the processes of second repayment source adjustment, remaining term adjustment, actual loan use adjustment, extension adjustment, and refinancing and restructuring adjustment, batch automatic extraction of debt characteristics, and the intersection of each process constraint is used as the final constraint score of the debt. If the intersection is empty, the upper limit of the set is taken as the final constraint.

[0086] S106: Considering the mutual influence between different debts of the same customer, further revise the system's initial classification results, and make batch cross-adjustment processes for customers. For example, when the same customer has suspicious or loss-making debts in a certain clearing month, the classification level of other businesses of the same customer should also be lowered accordingly, and the specific lowering level is controlled by the parameter module. First, retrieve the user's basic data and debt data and other data information, and for all the customer's debts that have completed the system's initial classification, determine whether these debts are all non-performing or all impaired.

[0087] S107: If all the debts that the customer has initially classified are non-performing or impaired, batch classification will be carried out according to the single contract, and the classification result will be adjusted to the worst classification result (including asset quality classification and risk classification) among all debts under the contract to ensure that the classification results of debts under the same contract are consistent.

[0088] S108: If the client has completed the initial classification of debts, and the initial classification results are not all non-performing or impaired, continue to determine whether the client's debts are completely non-performing or impaired.

[0089] S109: If the client has completed the initial classification of debts, and the initial classification results do not include any non-performing or impaired debts, the low-risk debts among them will not be subject to the client rule adjustment. They only need to be classified in batches according to a single contract to ensure that the classification results of debts under the same contract are consistent and the worst classification results. The asset classification level of low-risk debts is allowed to be inconsistent with the classification level of debts under other contracts of the client.

[0090] S110: If the customer has completed the initial classification of debts, and some of the initial classification results are classified as non-performing or impaired, but not all of them are classified as non-performing or impaired, then all debts of the customer are batch classified by single customer to ensure that the classification results of debts for the same customer are consistent and are the worst classification results among all debts of the customer.

[0091] S111: After the above judgment processes of S106, S107, S108, S109 and S110, after considering the mutual influence between the customer's credit assets, the worst adjustment score and grade among all adjustment results are output as the cross-adjustment result, and the system batch cross-adjustment is completed.

[0092] S112: The handling personnel shall judge whether to agree with the system classification results based on their own understanding of the actual situation of the customers and debts, combined with the results obtained from the batch classification and cross-adjustment of the system model. If they agree with the system classification results, they can confirm the results and proceed with the subsequent processes.

[0093] S113: If the handling personnel do not agree with the system classification results after cross-adjustment, they can manually intervene to adjust the system classification results and confirm the results, provided that the conditions are met.

[0094] S114: The handling personnel will submit the confirmed system classification results in batches for approval.

[0095] S115: Reviewers conduct batch reviews of the authenticity, validity, accuracy, and compliance of the client's asset classification results. If there are objections to the results of the previous process, or if other emergencies occur after the initial classification and the classification results must be adjusted, the classification results can be adjusted in batches during the review process (the relevant processing logic and technical implementation are the same as cross-adjustment).

[0096] S116: The reviewer conducts batch review of the client's asset classification results. During the review process, the classification results can be adjusted in batches (the relevant processing logic and technical implementation are the same as cross-adjustment) to complete the approval process.

[0097] S117: The classified debt data of the classification results are tested for transaction risks through a scoring model to measure the user's default risk and the transaction risk of the debt. If there are abnormal transaction risks, the abnormal information can be fed back to the management personnel.

[0098] Based on the same principle, this embodiment also discloses a financing risk monitoring device based on credit asset batch classification. Figure 6 As shown, in this embodiment, the device includes a debt initial classification module 11, a cross adjustment module 12 and an abnormality monitoring module 13.

[0099] The debt initial classification module 11 is used to perform initial classification of debts in batches on the user's basic data and debt data.

[0100] The cross adjustment module 12 is used to perform batch cross adjustment on the debt item data after batch initial classification to obtain classified debt item data.

[0101] The abnormality monitoring module 13 is used to determine whether the classified debt data has abnormal risks, and if so, to feedback abnormal information to the management personnel.

[0102] The financing risk monitoring method based on the batch classification of credit assets of the present invention performs batch initial classification of debts on the user's basic data and debt data, performs batch cross-adjustment on the debt data after batch initial classification to obtain classified debt data, and determines whether the classified debt data has abnormal risks. If so, abnormal information is fed back to the management personnel. It can be seen that the present invention automatically performs batch initial classification and batch cross-adjustment of debts, which can effectively avoid the problem of inconsistent classification results under the same customer or contract due to subjective or objective operational reasons of staff when classifying by single assets in the traditional way, reduce operational risks, and at the same time, financing risk monitoring improves the classification efficiency, asset security and risk monitoring capabilities of credit assets.

[0103] In a preferred embodiment, Figure 7 As shown, the device further includes a data processing module 10. The data processing module 10 is used to obtain the customer information and debt information stored in the transaction system before the basic data and debt data of the user are initially divided into debt batches. The external data information is obtained from the external system through an interface. The customer information, debt information and the external data information are decomposed and packaged to obtain the basic data and debt data of the user.

[0104] Specifically, when a customer conducts asset transactions through the trading system, the trading system stores the customer's customer information, debt information and other data information. Furthermore, in order to ensure the comprehensiveness of user data, external data information can be obtained from the external system through the interface. The external data information may include the total amount of external guarantees, other banks' quality classification information and other external data information. The data from different data sources are decomposed, packaged, classified and integrated to obtain the user's basic data and debt data, so as to provide it to other processes such as credit asset initial classification for call and calculation.

[0105] In a preferred embodiment, the data processing module 10 is further used to determine the corresponding user's basic data and debt data according to the screening dimensions input by the staff before performing initial batch classification of the user's basic data and debt data, and the screening dimensions include at least one of the time dimension, branch dimension and user dimension.

[0106] Specifically, the relevant data required for asset classification can be loaded to provide data preparation for the subsequent initial asset classification. At the same time, the loaded data information will be provided and displayed to the staff as the basic situation. The loading method may include manual single customer data loading and batch loading of bank data. Among them, the user's basic data may include basic customer information such as credit rating, industry, credit usage and default probability, and debt data may include basic debt information such as business balance, remaining debt term, restructuring situation, issuance date, basic transaction background and overdue (arrears) situation, as well as repayment data information such as second repayment source information such as mortgage guarantee data and guarantor information, to provide data preparation for the subsequent initial asset classification.

[0107] Furthermore, in order to facilitate the monitoring, modification and adjustment of the staff, the asset classification initial classification page can be displayed to the staff, and the customers to be classified under the institution can be queried and displayed with time and branch as the dimensions. The query results will display the business classification overview of each customer of the designated institution in the designated clearing month in the form of a list, and the single-household information can also be directly queried with the customer as the dimension. The staff selects the corresponding customer in the list of customers to be classified to automatically obtain the loaded customer basic data and debt data and other data information.

[0108] In a preferred embodiment, the data processing module 10 is further used to receive a debt modification request from a staff member before performing a batch initial division of the user's basic data and debt data. Modify the debt information of the corresponding target debt in the debt data according to the debt modification request. Modify the debt information of other debts of the user to whom the target debt belongs according to the debt modification request.

[0109] Specifically, if the data information changes, the classified debt data can be manually modified to provide more accurate data preparation for subsequent initial classification. If a single customer involves multiple debts, when modifying the information of one debt, the information of other debts can be updated in batches to achieve adjustments in the customer dimension. For example, when the debt information of the guarantor information in multiple debts in the debt data is associated and unified, when the staff modifies the guarantor information of one debt, the guarantor information of other debts associated with the debt will be automatically updated.

[0110] In a preferred embodiment, the debt preliminary classification module 11 is specifically used to perform preliminary classification of debts in batches on the user's basic data and debt data based on the customer evaluation index system and the debt evaluation index system through the preliminary classification model.

[0111] Specifically, the initial classification model can be used to measure the customer's default risk and the specific transaction risk of the debt. The two-step classification model of "customer classification + debt adjustment" is mainly adopted to perform qualitative and quantitative scoring on the classified debts, so as to obtain the corresponding twelve-level classification and three-stage classification results. Among them, the customer evaluation index system in the initial classification model mainly automatically evaluates the customer's behavior characteristics and calculates the customer classification result score through the processes of loss loan prediction, low-risk business prediction, loan commitment part initial classification, off-balance sheet business prediction, country risk adjustment, customer scale and economic composition adjustment, repayment willingness adjustment, customer status adjustment, credit line usage adjustment, etc. The debt evaluation index system adjustment is mainly based on the customer classification result score, through the processes of second repayment source adjustment, remaining term adjustment, actual loan use adjustment, extension adjustment and refinancing and restructuring adjustment, etc., to automatically extract debt characteristics in batches, and use the intersection of each process constraint as the final constraint score of the debt. If the intersection is empty, the upper limit of the set is taken as the final constraint.

[0112] In a preferred embodiment, the cross adjustment module 12 is specifically used for batch classification according to a single contract if all the debt data after the initial batch classification are non-performing or all impaired, and the classification result obtained is the worst classification result among all the debt data under the contract. If the debt data after the initial batch classification are not all non-performing or all impaired, it is determined whether the customer's debts are completely non-performing or impaired. If so, batch classification is performed according to a single contract to ensure that the classification results of debts under the same contract are consistent and the worst classification results; if not, all debts of the customer are batch classified according to a single customer to ensure that the classification results of debts under the same customer are consistent and the worst classification results among all debts of the customer.

[0113] Specifically, considering the mutual influence between different debts under the same customer, the system's initial classification results are further corrected to form a batch cross-adjustment process for customers. If all the debts that the customer has completed initial classification are non-performing or all are impaired, batch classification is performed according to a single contract, and the classification results are adjusted to the worst classification results (including asset quality classification and risk classification) among all debts under the contract to ensure that the classification results of debts under the same contract are consistent. If the customer has completed initial classification of debts, and the initial classification results are not all non-performing or all are impaired, then continue to determine whether the customer's debts are completely non-performing or impaired. If the customer has completed initial classification of debts, and the initial classification results do not completely contain non-performing or impaired debts, the low-risk debts among them will not participate in the customer rule adjustment. It is only necessary to perform batch classification according to a single contract to ensure that the classification results of debts under the same contract are consistent and the worst classification results are obtained. The asset classification level of low-risk debts is allowed to be inconsistent with the classification level of debts under other contracts of the customer. If the customer has completed the initial classification of debts, and the initial classification results include non-performing or impaired debts, but not all of them are non-performing or impaired debts, all debts of the customer will be classified in batches according to the single customer to ensure that the classification results of debts under the same customer are consistent and the worst classification results among all debts of the customer. After the above judgment process, considering the mutual influence between customer credit assets, the worst adjustment score and grade among all adjustment results are output as the cross-adjustment result, and the system batch cross-adjustment is completed.

[0114] In order to ensure the accuracy of the batch cross-adjustment results, the classification results of the batch cross-adjustment can be fed back to the staff so that the staff can confirm the classification approval results and review the authenticity, validity, accuracy and compliance of the classification results. If there is any objection to the results, or other emergencies occur after the initial classification and the classification results must be adjusted, relevant adjustments can be made to reduce operational risks. Specifically, the handling staff can judge whether to agree with the system classification results based on their own understanding of the actual situation of the customer and the debt, combined with the results obtained by the batch classification and cross-adjustment of the system model. If the system classification results are agreed, the results can be identified and the subsequent process can be carried out. If the handling staff does not recognize the system classification results after the cross-adjustment, the system classification results can be adjusted manually and identified under the premise that the conditions are met. The handling staff will submit the identified system classification results for approval in batches. The reviewer will conduct a batch review of the authenticity, validity, accuracy and compliance of the customer's asset classification results. If there is any objection to the results identified in the previous process, or other emergencies occur after the initial classification and the classification results must be adjusted, the classification results can be adjusted again in batches during the review process. The reviewer conducts batch review of the client's asset classification results. During the review process, the classification results can be adjusted again in batches to complete the approval process.

[0115] In a preferred embodiment, the abnormal monitoring module 13 is specifically used to perform transaction risk detection on the classified debt item data of the classification result through a scoring model. If the detection shows that there is abnormal transaction risk, the abnormal information is fed back to the management personnel.

[0116] Specifically, by using the scoring model to detect the transaction risk of the classified debt data of the classification results, the user's default risk and the transaction risk of the debt can be measured. If there is an abnormal transaction risk, the abnormal information can be fed back to the management personnel, and the abnormal debt can be transferred for disposal and management to monitor the financing risk of the user's credit assets. Among them, the scoring model can be trained by machine learning or neural network and other technologies based on historical normal debt data, which will not be repeated here.

[0117] Since the principle of solving the problem by the device is similar to that of the above method, the implementation of the device can refer to the implementation of the method and will not be repeated here.

[0118] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer device, and specifically, the computer device may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0119] In a typical example, the computer device specifically includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method executed by the client as described above is implemented; or, when the processor executes the program, the method executed by the server as described above is implemented.

[0120] Reference below Figure 8 , which shows a schematic diagram of the structure of a computer device 600 suitable for implementing an embodiment of the present application.

[0121] like Figure 8 As shown, the computer device 600 includes a central processing unit (CPU) 601, which can perform various appropriate operations and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0122] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal feedback device (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 606 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed as needed as the storage section 608.

[0123] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 609, and / or installed from the removable medium 611.

[0124] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0125] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0129] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0130] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0131] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0132] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0133] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A financing risk monitoring method based on batch classification of credit assets, characterized in that: include: Carry out initial batch classification of debts based on the user's basic data and debt data; Perform batch cross-adjustment on the debt item data after batch initial classification to obtain classified debt item data; Determine whether the classified debt data has abnormal risks, and if so, provide feedback on the abnormal information to the management personnel; The initial batch classification of the user's basic data and debt data specifically includes: The user's basic data and debt data are classified in batches based on the customer evaluation index system and the debt evaluation index system through the initial classification model; among them, the customer evaluation index system in the initial classification model automatically evaluates customer behavior characteristics and calculates customer classification result scores through loss loan prediction, low-risk business prediction, loan commitment part initial classification, off-balance sheet business prediction, country risk adjustment, customer scale and economic composition adjustment, repayment willingness adjustment, customer status adjustment, and credit line usage adjustment; based on the customer classification result score, the debt evaluation index system automatically extracts debt characteristics in batches through the adjustment of the second repayment source, the remaining term, the actual loan purpose, the extension situation, and the refinancing and restructuring situation, and uses the intersection of each process constraint as the final constraint score of the debt. If the intersection is empty, the upper limit of the set is taken as the final constraint; The batch cross-adjustment of the debt item data after the batch initial classification to obtain the classified debt item data specifically includes: If all debt items after the initial batch classification are classified as non-performing or impaired, the batch classification is performed according to the single contract, and the classification result obtained is the worst classification result of all debt items under the contract; If the debt data after the initial batch classification is not all non-performing or impaired, then determine whether the customer's debts are completely non-performing or impaired. If so, then batch classification is performed according to the single contract to ensure that the debt classification results under the same contract are consistent and are the worst classification results; if not, then all the customer's debts are batch classified according to the single customer to ensure that the debt classification results for the same customer are consistent and are the worst classification results among all the customer's debts.

2. The financing risk monitoring method based on credit asset batch classification according to claim 1 is characterized in that: It further includes before the initial batch classification of the user's basic data and debt data: Obtain customer information and debt information stored by customers in the trading system; Obtain external data information from external systems through interfaces; The customer information, debt information and external data information are decomposed and packaged to obtain the basic data and debt data of the user.

3. The financing risk monitoring method based on credit asset batch classification according to claim 1 is characterized in that: It further includes before the initial batch classification of the user's basic data and debt data: The basic data and debt data of the corresponding user are determined according to the screening dimensions input by the staff, and the screening dimensions include at least one of the time dimension, the branch dimension and the user dimension.

4. The financing risk monitoring method based on credit asset batch classification according to claim 1 is characterized in that: It further includes before the initial batch classification of the user's basic data and debt data: Receive debt modification requests from staff; Modify the debt information of the corresponding target debt in the debt data according to the debt modification request; The debt information of other debts of the user to which the target debt belongs is modified according to the debt modification request.

5. The financing risk monitoring method based on credit asset batch classification according to claim 1 is characterized in that: Determining whether the classified debt data has abnormal risks specifically includes: Conduct transaction risk detection on classified debt data of classification results through scoring model; If abnormal transaction risks are detected, the abnormal information will be fed back to the management staff.

6. A financing risk monitoring device based on credit asset batch classification, characterized in that: include: The debt initial classification module is used to perform initial classification of debts in batches based on the user's basic data and debt data; The cross adjustment module is used to perform batch cross adjustment on the debt item data after batch initial classification to obtain classified debt item data; An abnormality monitoring module is used to determine whether the classified debt data has abnormal risks, and if so, to feedback abnormal information to management personnel; The debt initial classification module performs initial classification of debts in batches on the user's basic data and debt data, specifically including: The user's basic data and debt data are classified in batches based on the customer evaluation index system and the debt evaluation index system through the initial classification model; among them, the customer evaluation index system in the initial classification model automatically evaluates customer behavior characteristics and calculates customer classification result scores through loss loan prediction, low-risk business prediction, loan commitment part initial classification, off-balance sheet business prediction, country risk adjustment, customer scale and economic composition adjustment, repayment willingness adjustment, customer status adjustment, and credit line usage adjustment; based on the customer classification result score, the debt evaluation index system automatically extracts debt characteristics in batches through the adjustment of the second repayment source, the remaining term, the actual loan purpose, the extension situation, and the refinancing and restructuring situation, and uses the intersection of each process constraint as the final constraint score of the debt. If the intersection is empty, the upper limit of the set is taken as the final constraint; The cross adjustment module performs batch cross adjustment on the debt item data after batch initial classification to obtain classified debt item data, specifically including: If all debt items after the initial batch classification are classified as non-performing or impaired, the batch classification is performed according to the single contract, and the classification result obtained is the worst classification result of all debt items under the contract; If the debt data after the initial batch classification is not all non-performing or impaired, then determine whether the customer's debts are completely non-performing or impaired. If so, then batch classification is performed according to the single contract to ensure that the debt classification results under the same contract are consistent and are the worst classification results; if not, then all the customer's debts are batch classified according to the single customer to ensure that the debt classification results for the same customer are consistent and are the worst classification results among all the customer's debts.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images