Small sample credit assessment method and system based on machine learning, and medium

A credit evaluation and machine learning technology, applied in the field of machine learning, can solve problems such as difficulty in data acquisition, small amount of data, poor credit model prediction effect, etc., to reduce credit risk, improve efficiency and income, and avoid data leakage risk Effect

Inactive Publication Date: 2021-08-06
BEIJING UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The invention is used to solve the problem of poor prediction effect of credit model due to difficulty in data acquisition and small amount of data

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
  • Small sample credit assessment method and system based on machine learning, and medium
  • Small sample credit assessment method and system based on machine learning, and medium
  • Small sample credit assessment method and system based on machine learning, and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] The overall implementation steps of the embodiment of the present invention are mainly divided into five steps:

[0028] The first step: obtain data, that is, obtain the original data set T required for modeling, including the feature variable P1 and the target variable. Characteristic variables refer to the available data about small and micro enterprises, such as the basic information of the enterprise, financial information and information about the legal person of the enterprise, as well as characteristic dimensions such as macroeconomic information. The specific variables that can be selected are shown in Table 1 below (Table 1 Only some variables are listed, and other publicly available data in the industry can also be included in the characteristic variables); the target variable refers to the historical overdue situation of the enterprise, that is, the historical overdue situation of the enterprise in the credit company or bank, usually expressed as 0 / 1, The def...

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 discloses a small sample credit assessment method and system based on machine learning, and a medium. The method comprises the following steps: selecting small sample data to train a generative adversarial network model, generating identically distributed pseudo data by using a generative network after the network reaches Nash equilibrium, then combining the pseudo data with real data to generate an amplified sample, and training a machine learning model by using the amplified sample for credit assessment. According to the method and system, the problems of difficulty in data acquisition, small data volume, non-uniform sample distribution, high data price and the like encountered by small sample data modeling at the present stage are solved, and meanwhile, the problems of poor credit evaluation model performance and the like caused by small sample data are also solved.

Description

technical field [0001] The invention belongs to the field of machine learning, and particularly relates to a small sample credit evaluation method, system and medium based on machine learning. [0002] technical background [0003] The construction of the existing credit model not only requires enough feature variables, but also usually has certain requirements for the capacity of the modeling samples. The small sample size cannot meet the requirements of existing machine learning methods such as scorecards and XGBoost, which leads to the failure of the model to achieve the expected accuracy and the purpose of distinguishing risks well. At the same time, individuals and enterprises are paying more and more attention to the protection of information and data, which makes it more difficult to obtain real data samples. [0004] The existing small sample learning methods are mainly based on the information induction and repeated derivation of feature variables, and are mainly us...

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
IPC IPC(8): G06Q40/02G06F16/21G06F16/215G06N3/04G06N3/08G06N20/00
CPCG06F16/215G06F16/212G06N20/00G06N3/04G06N3/08G06Q40/03
Inventor 刘海滨李健
Owner BEIJING UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap