CN116188140A - Credit financial risk policy production method and device - Google Patents
Credit financial risk policy production method and device Download PDFInfo
- Publication number
- CN116188140A CN116188140A CN202310011094.1A CN202310011094A CN116188140A CN 116188140 A CN116188140 A CN 116188140A CN 202310011094 A CN202310011094 A CN 202310011094A CN 116188140 A CN116188140 A CN 116188140A
- Authority
- CN
- China
- Prior art keywords
- risk
- credit
- financial
- parameters
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention relates to the field of internet finance, and provides a credit finance risk strategy production method and device. The problem of low production efficiency of the traditional manual strategy is solved, and the strategy approval operation production and the updating of the iterative financial risk strategy rule set can be rapidly assisted. The scheme comprises the following steps: acquiring financial performance information of a loan application user from an internal database, and generating financial performance characteristics according to the financial performance information; acquiring the latest characteristic parameters of the credit user by a third party database; inputting the financial performance characteristics and the three-party characteristic parameters into a trained credit risk assessment model to form a user risk score; any combination of the financial performance characteristics, the three-party characteristic parameters or the user risk scores is carried out to obtain a combination set for generating a plurality of parameters; inputting the parameter combination set into a machine learning model to calculate and obtain risk parameters; and when the risk parameters accord with the threshold value, the combination set corresponding to the risk parameters is integrated into a credit financial risk strategy.
Description
Technical Field
The invention relates to the field of internet financial retail credit strategies, in particular to a credit financial risk strategy production method and device.
Background
The internet finance is characterized in that the interaction with the client is performed in a data form, and the cognition of the client is completely from data, so that the capability of integrating the internal and external data information of the client is of great importance. In small, distributed credit financial products, the expiration of most customers is not usually a sudden loss of repayment capability, but rather a gradual process. This progressive process is a result of a combination of factors, both customer payback capacity and willingness to pay back. Therefore, it is important to effectively predict the future repayment behavior of the customer according to the historical repayment data, financial credit data and other third party auxiliary data of the customer. In addition, customers with high probability of default are excluded from the user borrowing process through various financial policy means.
In the formulation of credit financial risk policies, the current general technology is to conduct policy production based on the personal information, personal behavior or credit characteristics of the user. In order to provide a more comprehensive understanding for the user, the user portrait is more accurate and has stronger timeliness, the financial institution generally introduces updated data information of the third party database on the basis of the information of the internal database, and then combines the internal data information and the third party data information to formulate a risk policy. At present, policy mining analysis is mainly performed on financial characteristic information of a user and data information of a third party in a manual mode, and because a large number of characteristics of each client are used as entry items for policy combination formulation in the application of the data of the third party, the efficiency of the manual production policy method is particularly low. There is therefore a need for a credit financial risk policy production method and apparatus.
Disclosure of Invention
The invention provides a credit financial risk policy production method and device. Financial performance features can be generated based on financial performance information derivatives within the hierarchy and analyzed and modeled based on the latest features introduced by the third party database to generate a financial risk policy set. In the risk policy production method of the invention, an automatic analysis flow is established. The process can automatically combine the internal financial performance characteristics and the three-party data characteristics through an automatic device system, then parameter calculation and iteration are carried out on the combined set based on a model calculation system, and when the parameters meet the preset threshold requirements, the formed parameter combined set is finally produced into the financial risk strategy set. And if the newly added financial performance characteristics or the newly added third-party characteristics exist, automatically re-incorporating the updated characteristics based on the flow, performing parameter calculation, and combining the updated financial risk policy set with the existing parameters when the calculation meets the parameter threshold.
The invention aims to solve the problem of low production efficiency of the traditional manual strategy, and can quickly help the strategy approval operation to produce and update the iterative financial risk strategy rule set.
The invention provides a credit finance strategy production method, which mainly comprises the following steps:
step 1, acquiring financial performance information of a lending user from an internal database, and generating financial performance characteristics according to the financial performance information;
step 2, obtaining the latest characteristic parameters of the credit user by a third party database;
step 3, inputting financial performance characteristics or three-party characteristic parameters into a trained credit risk assessment model to obtain a user risk score;
step 4, the financial performance characteristics, or the characteristic parameters acquired by the three parties, or the user risk scores are arbitrarily combined to obtain a combination set of a plurality of parameters;
step 5, inputting the parameter combination set into a machine model to calculate and obtain risk parameters;
and 6, when the risk parameters accord with the threshold value, the combination sets corresponding to the risk parameters are integrated into a credit financial risk strategy.
In the above method, step 1 includes the steps of:
step 1.1, automatically acquiring financial performance information of a user from an internal database, wherein the financial performance information comprises names, identification numbers, mobile phone numbers, sexes, ages, professions, marital states, academies, loan balances and overdue times;
and step 1.2, generating financial performance characteristics based on the performance information acquired in the step 1.1.
In the above method, step 2 includes the steps of:
step 2.1, acquiring loan application user information from a third party database;
step 2.2, performing cleaning treatment to obtain the latest characteristic parameters;
and 2.3, eliminating a plurality of latest characteristic parameters of the user irrelevant to the credit risk, wherein the characteristic parameters comprise names, identification numbers, mobile phone numbers, current overdue states of outstanding loans, current overdue states of outstanding credit cards, the number of check-up queries of recent month credit, the number of check-up queries of recent three months, the number of recent month multi-head application institutions, the number of recent three month multi-head application institutions, the third party anti-fraud score and the third party credit score.
In the above method, step 3 includes the steps of:
step 3.1, inputting financial performance characteristics or three-party characteristic parameters into a trained credit risk assessment model for calculation;
and 3.2, forming at least one user risk score when the model score results meet the model condition.
In the above method, step 4 includes the steps of:
step 4.1, at least one financial performance characteristic is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.2, at least one three-party characteristic parameter is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.3, at least one user risk score is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.4, carrying out random combination on at least one of the financial performance characteristics, or at least one of the three-party characteristic parameters, or at least one of the user risk scores to produce a combined set of a plurality of parameters.
In the above method, step 5 includes the steps of:
step 5.1, inputting each parameter set into a machine model;
step 5.2, the machine learning model carries out iterative computation based on a plurality of parameters in a combined parameter set;
and 5.3, generating risk parameters of the combined parameter set when the iterative calculation meets the preset condition.
In the above method, step 6 includes the steps of:
step 6.1, determining to generate a credit financial risk sub-strategy when the risk parameter accords with a threshold value;
and 6.2, summarizing and sorting the generated multiple credit financial risk sub-strategies into a complete credit financial risk strategy.
The invention provides a financial strategy production device, which comprises: the internal data service module is used for preprocessing, calculating and processing the internal user data information to obtain at least one financial performance characteristic; the three-party data service module is used for acquiring at least one latest characteristic parameter of a user from a third-party data manufacturer; the credit model module is used for calculating financial performance characteristics or three-party characteristic parameters to form a user risk score; the feature parameter combination module is used for arbitrarily combining the financial performance features, the three-party feature parameters or the user risk scores to obtain a combination set for generating a plurality of parameters; the machine learning model module is used for inputting at least one parameter combination set into the machine model to calculate and acquire risk parameters; and the risk policy module is used for summarizing and integrating the combination set corresponding to the risk parameter into a credit financial risk policy when the risk parameter accords with the threshold value.
The beneficial effects of the invention are as follows:
1. the invention can automatically acquire the financial performance information of the loan application user in the system and generate the financial performance characteristics through the performance information.
2. The invention can automatically introduce the third party database, update and clean the information of the user, and keep the latest updated parameter characteristics.
3. The invention can input the financial performance characteristics or the three-party characteristic parameters into the existing credit risk model to calculate to obtain the user risk score, thereby helping to obtain the characteristic information with more comprehensive dimensions.
4. The invention can autonomously select at least one of the financial performance characteristics, or at least one of the three-way characteristic parameters, or at least one of the user risk scores to be arbitrarily combined to produce a combined set of a plurality of parameters.
5. According to the invention, a machine learning algorithm is introduced, each risk combination set is calculated and iterated to produce risk parameters, and corresponding risk sub-strategies can be produced when the risk parameters meet preset requirements.
6. The invention has obvious advantages in effect and speed in rule production, and provides a new method for generating credit financial risk strategies.
Drawings
The present invention will be described in detail with reference to fig. 1 and 2.
Fig. 1 shows a method for producing a financial strategy according to the present invention.
FIG. 2, the present invention establishes a set of automated strategy production facilities.
Detailed Description
In order to make the technical solutions and advantages of the present invention more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application.
All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.
Referring to the schematic flow chart 1, a financial strategy production method comprises the following detailed steps:
step 1, acquiring financial performance information of a loan application user from an internal database, and generating financial performance characteristics according to the financial performance information;
step 2, obtaining the latest characteristic parameters of the credit user by a third party database; and acquiring loan application user information from a third party database, performing cleaning processing to acquire the latest characteristic parameters, and eliminating the plurality of user latest characteristic parameters irrelevant to credit risks.
Time | XXXX |
Name of name | XXXX |
Identification card number | XXXX |
Mobile phone number | XXXX |
The current overdue state of outstanding loan | XXXX |
The status of the current overdue status of the unfit credit card | XXXX |
Number of credit approval queries for the next month | XXXX |
The number of credit approval query times of nearly three months accounts | XXXX |
Multi-head application organization number of near one month | XXXX |
Multi-head application organization number of nearly three months | XXXX |
Third party anti-fraud score | XXXX |
Third party credit score | XXXX |
...... | ...... |
Step 3, inputting financial performance characteristics or three-party characteristic parameters into a credit risk model for calculation; and forming at least one user risk score when the model score results satisfy the model condition.
Time | XXXX |
Name of name | XXXX |
Identification card number | XXXX |
Mobile phone number | XXXX |
ID | XXXX |
User risk scoring | XXXX |
...... | ...... |
And 4, carrying out any combination on the financial performance characteristics, the three-party characteristic parameters or the user risk scores to obtain a combined set for generating a plurality of parameters. And (3) any combination of at least one financial performance characteristic to obtain a combined set for generating a plurality of parameters, or any combination of at least one three-way characteristic parameter to obtain a combined set for generating a plurality of parameters, or any combination of at least one user risk score to obtain a combined set for generating a plurality of parameters, or any combination of at least one financial performance characteristic, or at least one three-way characteristic parameter, or at least one user risk score to generate a combined set for a plurality of parameters. For example: the financial performance characteristics are { A, B, C }, three-party characteristic parameters { X, Y, Z }, the risk scores { M, N }, and the finally obtained arbitrary combination set can be { A, B }, { A, X, Y }, { B, Y, Z }, { B, Z, M }, and the like.
Step 5, inputting the parameter combination set into a machine model to calculate and obtain risk parameters; each parameter set is input into a machine model, and the machine learning model performs iterative computation based on a plurality of parameters in the combined parameter set.
And step 6, determining to generate a credit financial risk sub-strategy when the risk parameter accords with a threshold value. And the generated multiple credit financial risk sub-strategies are integrated and tidied into a complete credit financial risk strategy.
Referring to the schematic flow chart of fig. 2, an automated policy production device is established, the policy production device 200 includes: an internal service module 201, a three-party data module 202, a credit model module 203, a feature parameter combination module 204, a machine learning model module 205, and a risk policy module 206.
The internal data service module 201 is used for preprocessing, calculating and processing the internal user data information to obtain at least one financial performance characteristic;
the third party data service module 202 is configured to obtain the latest feature parameters of the user from a third party data vendor, and reject the latest feature parameters of the plurality of users unrelated to credit risk;
a credit model module 203 for calculating financial performance characteristics or three-party characteristic parameters to form a user risk score;
the feature parameter combination module 204 is configured to perform any combination of the financial performance feature, the three-party feature parameter, or the user risk score to obtain a combined set that generates a plurality of parameters;
the machine learning model module 205 is configured to input at least one parameter combination set into a machine model to calculate and obtain risk parameters, and perform iterative calculation based on a plurality of parameters in the combination parameter set, and finally generate risk parameters of the combination parameter set when the iterative calculation meets a preset condition;
and the risk policy module 206 is configured to, when the risk parameter meets the threshold, sum up and integrate the combination set corresponding to the risk parameter into a credit financial risk sub-policy. And, multiple financial risk sub-policies may be cross-combined into the credit risk policy.
Claims (8)
1. The credit financial strategy production method is characterized by mainly comprising the following steps:
step 1, acquiring financial performance information of a lending user from an internal database, and generating financial performance characteristics according to the financial performance information;
step 2, obtaining the latest characteristic parameters of the credit user by a third party database;
step 3, inputting financial performance characteristics or three-party characteristic parameters into a trained credit risk assessment model to form a user risk score;
step 4, the financial performance characteristics, or the three-party acquired characteristic parameters, or the user risk scores are arbitrarily combined to generate a combined set of a plurality of parameters;
step 5, inputting the parameter combination set into a machine model to calculate and obtain risk parameters;
and 6, when the risk parameters accord with the threshold value, the combination sets corresponding to the risk parameters are integrated into a credit financial risk strategy.
2. The credit financial policy production method according to claim 1, wherein step 1 includes the steps of:
step 1.1, automatically acquiring financial performance information of a user from an internal database, wherein the financial performance information comprises names, identification numbers, mobile phone numbers, sexes, ages, professions, marital states, academies, loan balances and overdue times;
and step 1.2, generating financial performance characteristics based on the performance information acquired in the step 1.1.
3. The credit financial policy production method according to claim 1, wherein step 2 comprises the steps of:
step 2.1, acquiring loan application user information from a third party database;
step 2.2, performing cleaning treatment to obtain the latest characteristic parameters;
and 2.3, eliminating a plurality of latest characteristic parameters of the user irrelevant to the credit risk, wherein the characteristic parameters comprise names, identification numbers, mobile phone numbers, current overdue states of outstanding loans, current overdue states of outstanding credit cards, the number of check-up queries of recent month credit, the number of check-up queries of recent three months, the number of recent month multi-head application institutions, the number of recent three month multi-head application institutions, the third party anti-fraud score and the third party credit score.
4. The credit financial policy production method according to claim 1, wherein step 3 comprises the steps of:
step 3.1, inputting financial performance characteristics or three-party characteristic parameters into a trained credit risk assessment model for calculation;
and 3.2, forming at least one user risk score when the model score results meet the model condition.
5. The credit financial policy production method according to claim 1, wherein step 4 comprises the steps of:
step 4.1, at least one financial performance characteristic is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.2, at least one three-party characteristic parameter is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.3, at least one user risk score is arbitrarily combined to obtain a combined set for generating a plurality of parameters;
step 4.4, incorporating at least one of the financial performance characteristics, or at least one three-way characteristic parameter, or
At least one of said user risk scores is arbitrarily combined to produce a combined set of parameters.
6. The credit financial policy production method according to claim 1, wherein step 5 includes the steps of:
step 5.1, inputting each parameter set into a machine model;
step 5.2, the machine learning model carries out iterative computation based on a plurality of parameters in a combined parameter set;
and 5.3, generating risk parameters of the combined parameter set when the iterative calculation meets the preset condition.
7. The credit financial policy production method according to claim 1, wherein step 6 includes the steps of:
step 6.1, determining to generate a credit financial risk sub-strategy when the risk parameter accords with a threshold value;
and 6.2, summarizing and sorting the generated multiple credit financial risk sub-strategies into a complete credit financial risk strategy.
8. A financial policy production device, comprising:
an internal data service module: preprocessing, calculating and processing the internal user data information to obtain at least one financial performance characteristic;
three-party data service module: the method comprises the steps of obtaining at least one latest characteristic parameter of a user from a third party data manufacturer;
credit model module: the financial performance feature or the three-party feature parameters are calculated to form a user risk score;
and the characteristic parameter combination module is used for: any combination of the financial performance characteristics, the three-party characteristic parameters or the user risk scores is carried out to obtain a combination set for generating a plurality of parameters;
machine learning model module: for inputting the at least one parameter combination set into the machine model for computing an acquisition risk parameter;
risk policy module: and the method is used for summarizing and integrating the combination set corresponding to the risk parameter into a credit financial risk strategy when the risk parameter accords with the threshold value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310011094.1A CN116188140A (en) | 2023-01-04 | 2023-01-04 | Credit financial risk policy production method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310011094.1A CN116188140A (en) | 2023-01-04 | 2023-01-04 | Credit financial risk policy production method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116188140A true CN116188140A (en) | 2023-05-30 |
Family
ID=86441650
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310011094.1A Pending CN116188140A (en) | 2023-01-04 | 2023-01-04 | Credit financial risk policy production method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116188140A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117934161A (en) * | 2024-03-22 | 2024-04-26 | 杭银消费金融股份有限公司 | Method and system for evaluating clear-back payment in credit |
CN118037440A (en) * | 2024-04-09 | 2024-05-14 | 湖南三湘银行股份有限公司 | Trusted data processing method and system for comprehensive credit system |
-
2023
- 2023-01-04 CN CN202310011094.1A patent/CN116188140A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117934161A (en) * | 2024-03-22 | 2024-04-26 | 杭银消费金融股份有限公司 | Method and system for evaluating clear-back payment in credit |
CN118037440A (en) * | 2024-04-09 | 2024-05-14 | 湖南三湘银行股份有限公司 | Trusted data processing method and system for comprehensive credit system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116188140A (en) | Credit financial risk policy production method and device | |
Li et al. | Heterogeneous ensemble for default prediction of peer-to-peer lending in China | |
JP2018538587A (en) | Risk assessment method and system | |
CN109063931A (en) | A kind of model method for predicting freight logistics driver Default Probability | |
CN110909984B (en) | Business data processing model training method, business data processing method and device | |
CN108898476A (en) | A kind of loan customer credit-graded approach and device | |
CN113989019A (en) | Method, device, equipment and storage medium for identifying risks | |
CN112200656A (en) | On-line pre-approval method, device, medium and electronic equipment for house loan | |
CN112232950A (en) | Loan risk assessment method and device, equipment and computer-readable storage medium | |
US20130179255A1 (en) | Building and using an intelligent logical model of effectiveness of marketing actions | |
CN112950347A (en) | Resource data processing optimization method and device, storage medium and terminal | |
CN114723481A (en) | Data processing method and device, electronic equipment and storage medium | |
CN110738565A (en) | Real estate finance artificial intelligence composite wind control model based on data set | |
CN115759283A (en) | Model interpretation method and device, electronic equipment and storage medium | |
Alfat et al. | Feature Selection of Credit Score Factor Based on Smartphone Usage using MCFS | |
CN108009927A (en) | One B shareB methods of marking and platform | |
CN114240633A (en) | Credit risk assessment method, system, terminal device and storage medium | |
CN113256404A (en) | Data processing method and device | |
CN113240513A (en) | Method for determining user credit line and related device | |
CN112508608A (en) | Popularization activity configuration method, system, computer equipment and storage medium | |
Lu et al. | Predicting P2P lenders' decisions: the prospect theory approach | |
CN112561598B (en) | Customer loss prediction and retrieval method and system based on customer portrayal | |
Petchhan et al. | Toward Project Success Forecasting in Reward-based Crowdfunding through Wide-and-Deep Computational Learning | |
US20230342351A1 (en) | Change management process for identifying inconsistencies for improved processing efficiency | |
Tao et al. | Credit risk assessment of P2P lending borrowers based on SVM |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |