Method and apparatus for judging loss trend of user groups

Abstract

The invention discloses a method and an apparatus for judging a loss trend of user groups, relates to the technical field of computers, and mainly aims to judge a trend probability of user group loss for purchase behavior data of internet e-commerce users. According to the main technical scheme, the method comprises the steps of obtaining browsing behavior data of product purchase users, wherein the browsing behavior data is data information generated when the users perform product browsing; marking the users corresponding to the browsing behavior data with tags according to preset data eigenvalues, wherein the data eigenvalues are values of behavior feature fields in the browsing behavior data; classifying the browsing behavior data by utilizing the tags to obtain a user group set; and according to a ratio of the purchase behavior data of the user group set to the tags, calculating the loss trend of the user group set. The method and the apparatus are mainly used for predicting the loss trend of the user groups.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for judging the loss tendency of user groups. Background technique [0002] With the development of computer technology and the advent of the era of big data, various companies pay more and more attention to the retention of users, hoping to predict the loss of users. It is important for companies to identify users they already have who are at risk of churn early on, because if such users are identified early, steps can be taken to prevent churn. However, in the field of e-commerce, there is still a lack of clear and effective prediction mechanism for predicting the loss tendency of certain types of users. [0003] At present, the prediction algorithm for user churn does not clearly target the model algorithm for users on the e-commerce platform, so the current prediction algorithm mainly uses data sources such as user demographic information and us...

AI Technical Summary

Problems solved by technology

However, in the field of e-commerce, there is still a lack of clear and effective prediction mechanism for predicting the churn tendency of certain types of users.

[0003] At present, the prediction algorithm for user churn does not clearly target the model algorithm for users on the e-commerce platform, so the current prediction algorithm mainly uses data sources such as user demographic information and user behavior information as factors for user model construction

In addition, the current prediction model does not classify users according to Internet access data such as user purchase behavior data when predicting the churn tendency of users. It is predicted that e-commerce merchants will not be able to give feedback based on the user characteristics and preference models of different users after knowing the churn tendency

Images