The Research Of ZH Mobile Customer Classification Based On The Bayesian Classifier

With the development of China's telecommunications industry,the market have entered a new era of competition in the field of communication,the number of users is becoming saturated,more competition in service than a product of the competition.Guangdong Mobile,a company based in the area of customer service rated main line,introduced the current system of customer service.In addition,with the mobile industry reforms,competition,customer segmentation work will have a direct bearing telecommunication improvement of enterprises' core competence,decides its future development,but these issues have not been a good solution.How to reasonably segment customers according to their consumption behavior,to different customers with personalized service appropriate to their characteristics,create value for customers while achieving the maximization of corporate profits has become a telecommunications operator concerned an important issue.Discussed in the current situation,the shortcomings of service systems,and have put forward is established based on the value of a hierarchical classification of service management system,to satisfy the service under the current competitive situation and market demand.Present in the fledgling mobile communication operators in the field of customer management.Individual operators in the provinces in the area of mobile customer credit management a preliminary exploration and experiment.Based on the ZH research on mobile customer service management,rely on operating system data using cluster analysis on the State of Mobile Prepaid users graded model to customer classification based on Bayesian classifier.Override user wide model is simple and easy to implement,user identification feature.Finally,mobile tariff package on the basis of analysis of the current situation,given for category customer marketing advice.This thesis research on customer segmentation for mobile communication Enterprise provides effective way,as well as subscription packages analysis in depth laid a favorable Foundation.