Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Enterprise dishonest probability prediction method and system based on LightGBM

A probability prediction and enterprise technology, applied in the field of machine learning, can solve the problems of enterprise dishonesty probability prediction, etc., to improve the ability to prevent financing risks, increase and reduce the non-performing rate, and improve the effect of fraud prevention

Active Publication Date: 2019-11-05
UNIV OF SCI & TECH BEIJING
View PDF12 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is currently no method to predict the probability of corporate dishonesty and to accurately identify whether an enterprise will appear dishonest

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
  • Enterprise dishonest probability prediction method and system based on LightGBM
  • Enterprise dishonest probability prediction method and system based on LightGBM
  • Enterprise dishonest probability prediction method and system based on LightGBM

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0041] like figure 1 As shown, this embodiment provides a method for predicting the probability of enterprise dishonesty based on LightGBM, centering on the enterprise, and conducting business demand analysis and business demand understanding around the enterprise's reputation behavior footprint information in various aspects, and constructing a training data set. Divide all data into training set and test set, and perform preprocessing and feature engineering on the training set and test set respectively; use the training set to complete the development and design of the big data algorithm model, and use the fusion model to realize the detection of corporate dishonesty Accurate assessment. Its value lies in the use of algorithmic models to provide new ideas for solving corporate financing problems and improve the ability to prevent financing risks. Specifically, the step flow of the prediction method of this embodiment is as follows figure 2 shown, including:

[0042] S10...

no. 2 example

[0074] This embodiment provides a LightGBM-based enterprise dishonesty probability prediction system, which LightGBM-based enterprise dishonesty probability prediction system includes:

[0075] The first feature set building module is used to obtain the enterprise reputation behavior footprint information data set, construct the training data set, and perform preprocessing and feature extraction on the training data set, and construct the first feature set;

[0076] The first LightGBM model building block is used for training based on the first feature set using the LightGBM model to obtain the first LightGBM model;

[0077] The second feature set building module is used for training based on the first feature set, using three models of XGBoost, CatBoost, and LightGBM, and extracting the first 30 features of each model sorted by feature importance to construct the second feature set;

[0078] The second LightGBM model building block is used for training with the LightGBM model...

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 provides an enterprise dishonest probability prediction method and system based on LightGBM. The method comprises the following steps: analyzing and understanding credit behavior footprint information left by an enterprise in each aspect; preprocessing the data, carrying out further feature engineering on the existing data feature dimension in combination with service requirements; and then, reducing the dimension of the feature by using a feature selection and feature dimension reduction related method, learning data by using a machine learning model taking LightGBM as a main part, and obtaining a probability risk value of enterprise trust loss and a classification of whether the enterprise trust loss occurs or not by using the trained model. According to the technical scheme, the capabilities of preventing fraud and reducing the reject ratio of financial institutions can be further improved; accurate identification of whether an enterprise is credible is realized, problems of enterprise financing and credit evaluation are solved, financing risk prevention capability can be effectively improved, and the method can be widely applied to the field of enterprise loan auditing and enterprise social credit evaluation of banks.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a method for predicting the probability of enterprise dishonesty based on LightGBM. Background technique [0002] Credit is the foundation of the entire society, and all economic activities in market transactions are closely related to credit. At present, my country's enterprises are in the stage of rapid development, their influence is gradually expanding, and they have gradually become an important driving force for social and economic development. Therefore, it is imminent to strengthen the risk management and processing capabilities of the financing market, reduce the financing risk of enterprises, promote the development of the financing market, and establish a sound financing risk evaluation system; among them, accurately predict the probability of corporate dishonesty, and realize the accurate identification of whether the enterprise will appear dishonest , is th...

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): G06Q10/04G06K9/62G06Q10/06G06Q40/02
CPCG06Q10/04G06Q10/0639G06Q40/03G06F18/2135G06F18/214Y02P90/30
Inventor 阿孜古丽赵伟康谢永红张德政孙义栗辉孙宏飞
Owner UNIV OF SCI & TECH BEIJING
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com