Under-sampling classification integration method and device for credit scoring and storage medium

Abstract

The invention provides an under-sampling classification integration method and device for credit scoring and a storage medium. The method comprises the steps of obtaining a user training set, and dividing sample data in the training set into a majority class data set and a minority class data set; randomly undersampling k majority class data subsets from the majority class data set by using an undersampling algorithm, wherein each majority class data subset comprises a majority class data subset of n first data samples, and mn first data samples left after each time of undersampling form k pure majority class data subsets; mixing the k majority class data subsets with second data samples in the minority class data set to form k balanced data subsets; learning k CART tree dichotomy base classifiers by using the k balanced data subsets; utilizing the k pure majority class data subsets to learn k OnClassSVM classification base classifiers; and integrating the base classifiers through a bagging algorithm to output a final result. The problem of data imbalance in credit scoring is solved, and data samples are fully utilized to improve the classification performance.

Description

technical field [0001] The invention relates to the technical field of financial risk control, in particular to an undersampling classification integration method, device and storage medium for credit scoring. Background technique [0002] In credit loans, it is very important to evaluate the creditworthiness of loan applicants. Predicting the creditworthiness of lenders to decide whether to provide funds to borrowers has become a key issue in credit scoring. In credit, the number of instances of non-default class is far more than that of default class, showing the problem of class imbalance. In credit, there is a serious imbalance between the number of borrowers defaulting and the number of non-defaulting. Efficiently predicting credit risk from unbalanced datasets is difficult because unbalanced data affects the ability of classification models to distinguish good borrowers from potential defaulters. Traditional classification algorithms tend to favor the majority and de...

AI Technical Summary

Problems solved by technology

[0005] In view of the above-mentioned technical problems, the main purpose of the present invention is to provide an under-sampling classification integration method, device and storage medium for credit scoring. Improve classification performance by modifying or integrating existing algorithms

Images