The Customer Channel Preferences Analysis Based On Logistic Regression Model

With the rapid development of new technology and3G.4G license issuance, the threeoperators are involved in the whole business. The business enterprises, campuses, largepassenger of core markets have become intense. Faced with such a competitiveenvironment, operators have begun to focus on marketing, channel marketing.Data mining can analyze historical data telecommunications customers and find thedata in the underlying rules and patterns, and provide scientific reference views fordeveloping marketing strategies for telecom operators, business strategy.This Papermakes a brief introduction., the basic concepts of data mining to analyze the problem,modeling processes, and associated algorithms.Customer channels division is part of the customer relationship management.Toachieve customer channel preferences forecasts for the customer is the operator of the rawdata, using data mining techniques to further the basic customer information and otherdata.A combination of behavioral characteristics is analyzed.We summarized some of thebehavioral characteristics of customers, and through these characteristics, using datamining theory and methods, and establish channels of preference prediction model,In this paper, we study the mobile customer channel preference by customer research.It shows that lack of information about customer channel preferences is the main reasonand build the customer preference channel model is necessary. By using customerrelationship theory, statistical analysis theory, data mining theory, we build a customerchannel preference model. We use logistic regression and cluster analysis approach to dealwith customer preference data. The result provides a strong support to improve service.