User classification method and device, electronic equipment and storage medium

Abstract

The invention provides a user classification method and device, electronic equipment and a storage medium, and belongs to the technical field of computers. The method comprises the steps of obtaining multiple pieces of user behavior information corresponding to a to-be-classified user and a user feature variable corresponding to each piece of user behavior information; converting the user feature variables into first modulo variables according to preset conversion rules; according to a user behavior corresponding to the user behavior information, inputting each first modulo variable into a corresponding sub-model in a behavior classification model, so that each sub-model scores the input first modulo variable to obtain a plurality of behavior feature scores; performing binning on the plurality of behavior feature scores, and calculating a WOE value corresponding to each binning; and inputting the WOE value corresponding to each sub-box into a score card model to obtain a user category of the to-be-classified user, so that the evaluation deviation of user behaviors in different periods is relatively small, and the repayment willingness of the user is effectively evaluated.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a user classification method, device, electronic equipment and storage medium. Background technique [0002] Most of the existing user classification methods are based on user basic information, user borrowing behavior and other data to establish a scorecard model, and classify the user's repayment type according to the scorecard score. User information variables are processed before modeling, including screening and binning: the screening is mostly based on the IV value of the variable, and the binning is mostly based on equal-frequency binning, equidistant binning, or WOE binning. Modeling is mostly based on relatively simple machine learning models such as logistic regression and decision trees. However, the output results of these variable processing methods and model building methods are relatively coarse, and there is a large deviation in the classification of us...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiments of the present invention is to provide a user classification method, device, electronic equipment, and storage medium to solve the problem that the result granularity output by the above-mentioned variable processing method and model building method is relatively coarse, and there is a large deviation in user behavior classification in different periods question

Images