Unlock instant, AI-driven research and patent intelligence for your innovation.

Credit risk assessment method based on time sequence deep learning and legal document information

A technology of deep learning and risk assessment, applied in the field of credit risk assessment based on time-series deep learning and legal document information, can solve problems such as low efficiency, asymmetry, high non-performing rate and non-performing amount of banks, and improve work efficiency and service quality, reduce the non-performing rate and non-performing amount, and improve the effect of forecasting accuracy

Pending Publication Date: 2022-05-20
RENMIN UNIVERSITY OF CHINA
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Most of the traditional machine learning algorithms are based on basic statistical data, which cannot make good use of the chronological relationship of the data
Therefore, it is impossible to make full use of a large amount of transaction-level transaction data generated in the credit field to construct a risk assessment model. The prediction accuracy of the risk assessment model constructed by the existing technology is not high, and the bank's credit personnel will encounter insufficient information on the borrower during the pre-loan review. Such information asymmetry problems lead to high non-performing rate and non-performing amount in banks, and low banking efficiency

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Credit risk assessment method based on time sequence deep learning and legal document information
  • Credit risk assessment method based on time sequence deep learning and legal document information
  • Credit risk assessment method based on time sequence deep learning and legal document information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Embodiment one: if figure 1 As shown, the credit risk assessment model construction method based on time series deep learning and legal document information provided in this embodiment includes:

[0050] S1, such as figure 2 As shown, determine the optimal observation period and classify the judgment according to the loan applicant's litigation status and judgment outcome.

[0051] Specifically, in order to select the best observation period before extracting legal document information, this embodiment uses the existing common method of testing correlation chi-square test to test the correlation between the observation period and loan default, and find out the relationship between the default The observation period with the highest correlation.

[0052] In order to determine which judgments are effective in predicting credit risk, the legal document judgments in the selected observation period are divided into four categories: the type of judgment that the litigation...

Embodiment 2

[0123]Embodiment 2: The first embodiment above provides a credit risk assessment method based on time-series deep learning and legal document information. Correspondingly, this embodiment provides a credit risk assessment system based on time-series deep learning and legal document information. The system provided in this embodiment can implement the credit risk assessment method based on time-series deep learning and legal document information in Embodiment 1, and the system can be implemented by software, hardware, or a combination of software and hardware. For the convenience of description, when describing this embodiment, functions are divided into various units and described separately. Of course, the functions of each unit can be realized in one or more pieces of software and / or hardware during implementation. For example, the system may include integrated or separate functional modules or functional units to execute corresponding steps in the methods of the first embod...

Embodiment 3

[0130] Embodiment 3: This embodiment provides an electronic device corresponding to the credit risk assessment method based on time-series deep learning and legal document information provided in Embodiment 1. The electronic device can be an electronic device for a client, such as a mobile phone, Notebook computer, tablet computer, desktop computer etc., to carry out the method of embodiment one.

[0131] Such as Figure 7 As shown, the electronic device includes a processor, a memory, a communication interface, and a bus, and the processor, memory, and communication interface are connected through the bus to complete mutual communication. The bus may be an Industry Standard Architecture (ISA, Industry Standard Architecture) bus, a Peripheral Component Interconnect (PCI, Peripheral Component) bus, or an Extended Industry Standard Architecture (EISA, Extended Industry Standard Component) bus, and the like. A computer program that can run on the processor is stored in the memor...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a credit risk assessment method based on time sequence deep learning and legal document information, and the method comprises the steps: determining an optimal observation period, and classifying the judgment according to the litigation condition and judgment result of a loan applicant; crawling legal decision documents within a set time, configuring document entity extraction rules and dictionaries, and performing legal document entity extraction by adopting a rule-based extraction method; preprocessing the extracted legal instrument data, and performing event extraction on the legal instrument text information; selecting legal document features with strong predictive ability by using an RFE recursive feature selection method; and setting a mixed data set and training the LSTM model to obtain an evaluation model for credit risk evaluation. According to the method, early risk identification, early warning and early discovery can be realized, customer risk early warning can be initiated in time, the risk control quality is improved, a more accurate and more reliable basis is provided for bank anti-fraud application decisions, algorithm practice is promoted to enable user risk management, and the reject ratio and the defective rate of banks are effectively reduced.

Description

technical field [0001] The invention relates to a credit risk assessment method based on time series deep learning and legal document information, and relates to the field of computer science and technology. Background technique [0002] Currently, in the field of credit risk assessment, the developed financial transaction market generates a large amount of transaction-level transaction data containing time-series information all the time. In the face of a large amount of time-series transaction data, traditional machine learning models cannot solve the problem of gradient disappearance and gradient explosion in the long sequence training process, resulting in large-scale data not being fully mined and applied. In addition, legal judgment documents contain a wealth of information, which to a certain extent reflects the default risk of the borrower. The construction of the feature set of the previous credit risk assessment model often ignores the information of legal document...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/06G06Q40/02G06Q50/18G06F40/242G06F40/295G06K9/62G06N3/04G06N3/08
CPCG06Q10/0635G06Q50/18G06F40/295G06F40/242G06N3/049G06N3/08G06N3/045G06Q40/03G06F18/24323
Inventor 许伟杜玮王明明周宣晔
Owner RENMIN UNIVERSITY OF CHINA