Five-order system special post-loan management method and device based on big data analysis

Abstract

The invention provides a five-order system special post-loan management method and device based on big data analysis, and the method and device achieve the refined management and dynamic risk monitoring of special loans through the integration of bank internal data and Internet credit investigation data, the generation of multi-dimensional customer tags, and the combination of five-order system special classification and an intelligent post-loan management strategy. The device can effectively improve the risk control capability of banks in post-loan management, and is suitable for the field of financial science and technology.

Description

Technical Field

[0001] The present invention relates to the field of fintech, and particularly to a five-single special post-loan management method and device based on big data analysis, which are used to improve the risk control ability of banks in post-loan management, especially for the management of special loans (such as risky loans, loans with lost repayment ability, loans with lost limitation of action, early warning loans, litigation loans, etc.). Background Art

[0002] In the banking industry, post-loan management is an important link in risk control, especially for the management of special loans (such as loan in the name of another person, borrower's loss of repayment ability, loss of limitation of action, etc.). Traditional post-loan management methods rely on manual operations, which are inefficient and prone to missing key risk points. With the development of big data technology, banks can conduct in-depth analysis of customer behavior through multi-source data (such as bank internal data, Internet credit investigation data, etc.). However, existing post-loan management systems lack the ability of refined management and intelligent processing for special loans and cannot effectively handle complex risk scenarios.

[0003] In addition, most existing post-loan management systems rely on static rule engines and lack the ability of real-time monitoring and early warning of dynamic risks. Especially when facing special loans, it is difficult to quickly identify risks and take corresponding management measures. Therefore, there is an urgent need for a post-loan management device that can combine big data analysis and intelligent management to improve the efficiency and accuracy of banks in special loan management. Summary of the Invention

[0004] The purpose of the present invention is to provide a five-single special post-loan management method and device based on big data analysis. Through big data analysis technology, deep mining of customer data is carried out, combined with intelligent post-loan management strategies, to achieve refined management of special loans and dynamic risk monitoring, and improve the risk control ability of banks in post-loan management.

[0005] (I) Technical Solution

[0006] First, the data sources are integrated and analyzed, specifically including the following data:

[0007] Bank internal data: including the full amount of personal customer information, the full amount of corporate customer information, the increment of customer credit information, the full amount of personal guarantee information, the full amount of corporate guarantee information, the full amount of personal loan information, the full amount of corporate loan information, the full amount of personal written-off loan information, the full amount of corporate written-off loan information, the full amount of old written-off loan information, the full amount of personal replacement loan information, the full amount of corporate replacement loan information, the full amount of personal loan repayment plans, the full amount of corporate loan repayment plans, the increment of personal loan repayment records, the increment of corporate loan repayment records, customer deposit data, customer wealth management data, customer intermediate business data, customer acquiring transaction data, etc.

[0008] Internet credit investigation data: including high-risk lending behavior levels, risk litigation lists, length of time the mobile phone number has been in service, complex network fraud scores, risk assessment levels of false information, confidence in personal information, personal lending behavior assessment indices, changes in lending behavior trends, lending behavior preference degrees, consumer behavior assessment levels, device behavior assessment levels, payment behavior assessment levels, consumer behavior characteristics, device behavior characteristics, payment behavior characteristics, tracking of changes in lending behavior trends, tracking of lending behavior preference degrees, etc.

[0009] Furthermore, customer tags are defined, divided into 15 major categories and more than 300 tags, specifically as follows:

[0010] Borrower tags: such as normal, migrant worker, ill, deceased, family in difficulty, etc.

[0011] Production and operation situation tags: such as normal, change of business location, change of business project, change of legal representative, suspension of business, closure due to business failure, etc.

[0012] Repayment situation tags: such as normal repayment, ability to repay but no willingness to repay, no ability to repay but willing to repay, no ability to repay and no willingness to repay, etc.

[0013] Special guarantee situation tags: such as change of guarantor, change of collateral, invalid guarantee, guarantee by public officials, guarantee by internal employees, etc.

[0014] Contact tags: such as empty number, number has been reassigned, customer refuses to answer the phone, difficult to contact due to long-term absence, etc.

[0015] Litigation situation tags: such as customer requests litigation, has been transferred to a law firm, pending filing with the court, reached a pre-litigation settlement, etc.

[0016] Risk tags: such as suspected loan in another's name, suspected loan under a false name, borrower did not appear to sign, guarantor did not appear to sign, etc.

[0017] Specific tags: such as loan in another's name, loan under a false name, loan under a fake name, empty exhibition, empty loan, interest collection with loans, etc.

[0018] Customer type tags: such as farmers, self-employed individuals, government agencies, rural economic organizations, companies and enterprises, etc.

[0019] Guarantee tags: such as credit, guarantee, mortgage, pledge, etc.

[0020] Industry tags: such as agriculture, forestry, animal husbandry, fishery, mining, manufacturing, etc.

[0021] Loan tags: such as below 50,000 yuan, 50,000 - 300,000 yuan, 300,000 - 500,000 yuan, 500,000 - 1,000,000 yuan, etc.

[0022] Loan product labels: such as personal first-hand commercial housing loans, small business start-up loans, personal comprehensive consumption loans, etc.

[0023] Performance labels: such as normal status, overdue, five-grade non-performing, five-grade normal, etc.

[0024] Litigation status labels: such as less than 90 days of statute of limitations, less than 90 days of guarantee period, first instance case filed but not yet processed, etc.

[0025] Furthermore, classify the special loans under the five-single system and formulate post-loan management strategies. The specific classifications and corresponding management strategies are as follows:

[0026] Risk category: For high-risk loans such as loan taken out under a false name, loan taken out under a fictitious name, and invalid loan contract, the system pushes work orders to the risk department annually. The risk department can push the work orders to the customer manager for follow-up for a second time.

[0027] Category of loss of repayment ability: For loans where the borrower and the guarantor have died or lost full capacity for civil conduct, the system pushes work orders to the risk department and the customer manager annually. The customer manager needs to conduct on-site collection once a year.

[0028] Category of loss of statute of limitations: For loans where the statute of limitations has been lost or the guarantee period has expired, the system pushes work orders to the customer manager annually. The customer manager can push the work orders to the legal department for follow-up.

[0029] Early warning category: For loans where the borrower or the guarantor has the ability to repay but refuses to do so, the statute of limitations is less than 90 days, or the guarantee period is less than 90 days, the system pushes work orders to the legal department quarterly. The legal department can push the work orders to the customer manager for follow-up for a second time.

[0030] Litigation category: For loans where litigation has been filed, the system pushes work orders to the customer manager quarterly. The customer manager can push the work orders to the legal department for follow-up.

[0031] Finally, achieve intelligent post-loan management to realize real-time monitoring and early warning: Through big data analysis technology, real-time monitor the customer's repayment behavior, litigation status, guarantee situation, etc., and trigger risk warnings. Dynamic risk scoring: Based on the multi-dimensional labels of customers, dynamically calculate the risk scores of customers to help the bank quickly identify high-risk customers. Automated work order push: According to the special classification of customers, the system automatically generates and pushes work orders to relevant departments (such as the risk department, legal department, customer manager, etc.) to ensure that each special loan can be processed in a timely manner.

[0032] (2) Technical effects

[0033] The present invention is a method and device for special post-loan management under the five-single system based on big data analysis, which can achieve the following technical effects:

[0034] Implement refined post-loan management: Through multi-dimensional customer tags and the five-document classification system, achieve refined management and dynamic monitoring of special loans.

[0035] Implement intelligent risk warning: Based on big data analysis and dynamic risk scoring, monitor the repayment behavior and litigation status of customers in real time and trigger risk warnings.

[0036] Implement automated work order processing: The system automatically generates and pushes work orders to ensure that each special loan can be processed in a timely manner, improving the efficiency and accuracy of post-loan management.

[0037] Implement collaborative management among multiple departments: Through the work order push mechanism, achieve collaborative management among multiple departments such as the risk department, legal department, and customer managers, improving the overall efficiency of post-loan management. Description of the Drawings

[0038] Figure 1 : Architecture diagram of the five-document system for special post-loan management based on big data analysis.

[0039] Figure 2 : Definition and classification of customer tags.

[0040] Figure 3 : Five-document system special classification and post-loan management strategy flowchart.

[0041] Figure 4 : Intelligent post-loan management flowchart. Detailed Implementation Manner

[0042] The present invention will be further described in detail below in conjunction with specific embodiments, but it is not limited to the present invention.

[0043] First, integrate the data sources. By importing the data within the bank, specifically including the incremental query of loan occurrence details, the incremental query of customer information, the full query of guarantors (guarantee contracts), the full loan ledger, the full information of the written-off loan ledger, the full information of the old written-off ledger query, the non-performing loan replacement ledger, the full repayment plan of installment loans, the loan interest details, the inbound and outbound of collateral, the full information of collateral inbound and outbound, and the Internet credit data, integrate multi-dimensional data such as the personal information, loan information, repayment records, guarantee information, overdue information, written-off information, and collateral of customers from these original data.

[0044] Further, by adopting big data analysis technology, the customer data is deeply mined to generate multi-dimensional customer tags, and the customers are tagged. Before that, the tags are classified and defined first, and rules are set for each tag. If the previous data meets the set rules, the corresponding tag is given to the customer. In this embodiment, 15 types of tags are defined, namely borrower tag, production and operation situation tag, repayment situation tag, special guarantee situation tag, contact tag, litigation situation tag, risk tag, specific tag, customer type tag, guarantee tag, industry tag, loan tag, loan product tag, performance tag, litigation status, etc., with a total of nearly 300 tags.

[0045] Further, a five-single classification is set. In this embodiment, five types, namely risk category, loss of repayment ability category, loss of limitation of action category, early warning category, and litigation category, can be called the five-single system. Each type is set according to the customer's tags. For example, a risk-class loan means there is a loan in the name of another person or a false loan or an invalid loan contract, and the loan balance is greater than 0 and there is no litigation. In this way, a tag operation formula is configured for each category to screen customers and see which customers meet which category.

[0046] Further, different management strategies are set for various types of loans. The management strategy for the risk category is to push work orders to the risk department annually, and the risk department can push them to the customer manager for follow-up twice; the management strategy for the loss of repayment ability category is to push work orders to the risk department and the customer manager annually, and the customer manager needs to conduct on-site collection once a year; the management strategy for the loss of limitation of action category is to push work orders to the customer manager annually, and the customer manager can push the work orders to the legal department for follow-up; the management strategy for the early warning category is to push work orders to the legal department quarterly, and the legal department can push them to the customer manager for follow-up twice; the management strategy for the litigation category is to push work orders to the customer manager quarterly, and the customer manager can push the work orders to the legal department for follow-up.

[0047] Further, the device automatically generates and pushes work orders to relevant execution parties, such as multiple departments including the risk department, legal department, customer manager, outbound call center, law firm, etc., and external roles collaborate to process the work orders to ensure that each special loan can be processed in a timely manner.

[0048] Finally, by real-time monitoring the customer's repayment behavior, litigation status, guarantee situation, etc., a risk warning is triggered. Based on the multi-dimensional tags of the customer, the risk score of the customer is dynamically calculated to help the bank quickly identify high-risk customers.

[0049] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirits of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

[0050] The above describes the present invention and its implementation manners. Such description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. All in all, if those of ordinary skill in the art are inspired by it and, without departing from the gist of the present invention, creatively design structural manners and embodiments similar to the technical solution, they shall fall within the protection scope of the present invention.

Claims

1. A special post-loan management method based on big data analysis, characterized in that, It includes the following steps: Integrate in-bank data and Internet credit investigation data of the bank to generate multi-dimensional customer tags; According to the special classifications of customers (such as risk category, loss of repayment ability category, loss of limitation of action category, warning category, litigation category), the system automatically generates and pushes work orders to relevant departments; Trigger risk warnings by monitoring the repayment behavior, litigation status, guarantee situation, etc. of customers in real time; Based on the multi-dimensional tags of customers, dynamically calculate the risk scores of customers to help the bank quickly identify high-risk customers.

2. The method according to claim 1, wherein The customer tags include borrower tags, production and operation situation tags, repayment situation tags, special guarantee situation tags, contact tags, litigation situation tags, risk tags, specific tags, customer type tags, guarantee tags, industry tags, loan tags, loan product tags, performance tags, litigation status tags, etc.

3. The method according to claim 1, wherein The five-sheet system special classifications include risk category, loss of repayment ability category, loss of limitation of action category, warning category, litigation category.

4. The method according to claim 1, wherein The system automatically generates and pushes work orders to multiple departments such as the risk department, legal department, customer manager, etc., to ensure that each special loan can be processed in a timely manner.

5. A special post-loan management device based on big data analysis, characterized in that, It includes: A data integration module for integrating in-bank data and Internet credit investigation data of the bank to generate multi-dimensional customer tags; A work order generation module for automatically generating and pushing work orders to relevant departments according to the special classifications of customers; A risk warning module for triggering risk warnings by monitoring the repayment behavior, litigation status, guarantee situation, etc. of customers in real time; A risk scoring module for dynamically calculating the risk scores of customers based on the multi-dimensional tags of customers.

6. A work order execution module for executing work order task terminal devices.

7. The device according to claim 5, characterized in that, The work order generation module automatically generates and pushes work orders to multiple departments such as the risk department, legal department, customer manager, etc. according to the special classifications of customers (such as risk category, loss of repayment ability category, loss of limitation of action category, warning category, litigation category).

8. The device according to claim 5, characterized in that, The risk warning module triggers risk warnings by monitoring the repayment behavior, litigation status, guarantee situation, etc. of customers in real time.

9. The device according to claim 5, characterized in that, The risk scoring module dynamically calculates the risk scores of customers based on the multi-dimensional tags of customers to help the bank quickly identify high-risk customers.

10. The device according to claim 5, characterized in that, The work order execution module includes a customer manager mobile terminal, an outbound call center terminal, a department agency PC terminal, a law firm terminal, etc., and each terminal processes the work order tasks pushed to its own terminal.

Images