Feature selection optimization method and device and readable storage medium

Abstract

The invention discloses a feature selection optimization method and device and a readable storage medium, and the method comprises the steps: obtaining a sample binning result corresponding to each to-be-selected feature, and carrying out the feature selection of each to-be-selected feature based on each sample binning result, obtaining a candidate feature set which does not meet the requirements of a preset feature evaluation value and a target feature set which meets the requirements of the preset feature evaluation value; determining a target sample binning corresponding to each candidate feature according to each binning feature evaluation value of each candidate feature in the candidate feature set on the respective corresponding sample binning result; based on each target sample sub-box and the candidate feature set, constructing each target cross feature meeting the requirement of the preset feature evaluation value; and adding each target cross feature into the target feature set to obtain a target feature selection result. The technical problem that the feature selection effect is poor is solved.

Description

technical field [0001] The present application relates to the technical field of machine learning, and in particular to a feature selection optimization method, device and readable storage medium. Background technique [0002] With the rapid development of computer technology, the application of machine learning is becoming more and more extensive. In order to improve the efficiency and effect of machine model building, feature selection is usually required before modeling. At present, feature selection methods are usually based on However, since the calculation of IV value is related to the degree of dispersion of the distribution of feature values, the fairness of feature selection based on IV value will be affected by the degree of dispersion of distribution of feature values. , leading to low fairness of feature selection based on IV value, which reduces the effect of feature selection, and the features eliminated by feature selection will also lose some valuable modelin...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of this application is to provide a feature selection optimization method, device and readable storage medium, aiming to solve the technical problem of poor feature selection effect in the prior art

Images