Application Of Data Mining For Customer Subdivision Of Telecommunication Company

With constant aggravation of the telecommunication market competition, how to implement difference marketing and service to different customers group, how to subdivide and classify the active demand that has already become present telecommunication enterprises to the customer.This article has discussed the current situations of enterprise's customer of telecommunications subdivides and development trends of the work at present, and have analysed that excavates technology of the data to apply to the necessity and feasibility in telecommunication enterprises, is it excavate data cluster analysis apply to customer , telecommunication of enterprise subdivide and work to propose.Data Mining is the technique that refines the hidden, unknown but potential useful information and knowledge from huge mount of, incomplete, noisy, fuzzy and random business data.This text has described the basic theories in data warehouse and data mining; It is analysed that the main method used in data mining usually adopts at present; The K-Mean fast cluster's algorithm in cluster's analysis is deeply studied; Provide and apply it to modeling course that telecommunication customer subdivides.K-Mean fast cluster's algorithm is the most frequently used one in cluster's algorithm in the data mining, it calculates the distance between samples and initial central point, is adjusting the position of the central point constantly, make central point distance to get some sample and reach most short , realize " assemble " and " classify " of sample.Combine the management system and business mode of Chongqing mobile communication Co., Ltd. following the CRISP –DM standard data mining course, this text has expounded the fact in detail of Business understanding, Data understanding, Data preparation, Modeling, Evaluation, Deployment. At first , K-Mean fast cluster is proceed based on various kinds of consumer behaviors , then combining with market experience of business , divide them into tens of groups between whitch the behavior is enormous differenct and in whitch the behavior is little differenct.Finally,"CQMC intellectual marketing system" was developed based on telecommunication enterprise's customer subdivide model, in which B/S struct and software engineer theories ware used.It have realized the application of enterprise's customer of telecommunications subdivide,and help telecommunication operator's carrying on difference marketing to different customers group, improves the enterprise's key competitiveness of telecommunication...