Feature item processing method and device for credit investigation data, computer device

Abstract

The invention relates to a feature item processing method and device for credit investigation data, a computer device and a storage medium. The method comprises: sample data of credit investigation data are grouped; feature items are extracted from the sample data; an information value of each feature item is calculated; feature items with the information values higher than a preset value are extracted; information gain values and Gini coefficients of the extracted feature items are calculated; according to the information gain values and Gini coefficients, the feature items are screened to obtained the screened feature items; and on the basis of a stepwise regression algorithm, multicollinearity feature items among the screened feature items are rejected to obtain a final feature item. Because of calculation of the information gain values and the Gini coefficients, an effective feature item is selected out quickly; and the multicollinearity feature items among the feature items are removed by using the stepwise regression algorithm, so that the credit score becomes accurate.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method, device, computer equipment and storage medium for processing feature items of credit information data. Background technique [0002] With the continuous development of the market economy and the continuous increase of credit report data, there are more and more feature items in the data of credit report data. Feature items are known variables that have a significant distinguishing effect on the target variable. When scoring a customer's credit, it is usually necessary to select an effective feature item for modeling to score the customer's credit. The traditional processing method of characteristic items is to calculate the information value of the characteristic items, and select the characteristic items whose information value meets the standard for modeling to score the customer's credit. [0003] The traditional processing method of feature items is relat...

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to provide a feature item processing method, device, computer equipment and storage medium for the above-mentioned feature items that cannot be effectively mined and affect credit scoring.

Images