Predictive analysis of the loss of online shopping users based on data mining

Abstract

In the fast-changing Internet era, the advantages of e-commerce over traditional shopping models are becoming more and more obvious, and convenient and fast online shopping patterns are attracting more and more users. At the same time, large-scale transactions and demand between e-commerce competition is becoming increasingly fierce, inter-enterprise competition on the one hand to promote the development of e-commerce, at the same time, but also accelerate the survival of e-commerce. Enterprise competition has intensified, customers to the enterprise, has become the most important resource, how to attract customers and retain customers has become the focus of the enterprise, which also makes customer loss become the concern of many enterprises. E-commerce companies in order to ensure their own healthy development in the fierce competition market, not only to make their products attractive, but also in-depth understanding of user preferences and satisfaction, the user’s behavior characteristics of in-depth exploration. E-commerce user behavior instability is greater, the churn rate is high, then, can we find customers in time before the loss, while helping the marketing department to target the loss of customer base and develop appropriate marketing programs is an important work of the enterprise marketing department.It is an important work in the daily operation and management of e-commerce enterprises to predict the loss of users more accurately, to implement targeted retention strategies for users who are at greater risk of loss, and to reduce the churn rate. In these areas, data mining can help businesses. In this paper, data mining technology is applied to business analysis to predict the loss of Tmall users within a certain period of time, so as to implement retention strategy and reduce the churn rate.