Research On Customer Segmentation Model And Methods For Mobile Virtual Network Operator

With constant aggravation competition in the telecommunications market, how to implement the differentiated marketing and service for the different customer groups, it has already become the pressing demand of the present mobile virtual network operators that how to subdivide and classify to the customer. So research on customer segmentation model and methods for mobile virtual network operators, it can better to solve the real problem which customer segmentation for mobile virtual network operators, it is propitious for the mobile virtual network operators to know the customer composition, the characteristics of the customer groups and the characteristics of cost; Then develop the appropriate marketing policy and control operation status.This thesis discusses the necessity and feasibility of customer segmentation for present mobile virtual network operators, state the theory of related data mining, build customer segmentation model for mobile virtual network operators, and use cluster algorithm to implement customer segmentation.The main content of this article is the technology of the data mining to apply in the mobile virtual network operators of customer segmentation. First, introduce the generation and influence of the mobile virtual network operators, and the necessity and feasibility to set up the customer segmentation model; after that, to introduce the basic concepts of the data mining, basic tasks, implementation process and main algorithms, focuses on cluster algorithm. Then, build customer segmentation model with partitioning method. After that, combine the customer data and customer behavior data of the mobile virtual network operator ABC, use the K-means algorithm of SPSS Clementine data mining software to classify customers according to the GRISP-DM standard process. Last, develop the K-medoids algorithm with Visual Studio tool, compare and analyze the result based on these two algorithms.