The Research And Application Of Clustering Algorithm In The Classification Of The Telecom Customers

As the rapid development of social economy,informationization penetrates unceasingly and deeply.The advent and development of the information age,especially the development of Internet communications technology,makes the Global telecommunications consumer market developing into the 4G age.With the virtual operators taking part in,the competition of the three operators is becoming more and more intense.As a staff member of the operator,the author myself also feels that the demand of data mining increasingly strong for the company.Currently,the operators in our country are testing into the data mining field,because what they guard is a"gold mine" of information.How to use proper technology to mine business information to satisfy the growing demands of the users,is the prerequisite of achieving the ideals about "Corporation value and Customers' value growing together".This article is written from the perspective of a routine data carrier operation and maintenance personnel.Through daily experience and scientific methods,I establish a data model and analyze if by using cluster analysis.Analyze and put cluster analysis,an important field of Data mining,into real business decisions.The essence of cluster analysis is to gather and grouping the non-entity objects.Its main function is to gather and grouping the similar data and provide a appropriate data analysis solution for the operator.The article is written from the related technology that studying the relevant knowledge and background of the Customer Relationship Management System,Customers separation,data mining and Cluster analysis comprehensively.Analyzing and designing the telecommunications customers include the development of specific mining objectives,the use of analytical methods,and customer analysis of business understanding.After finishing the data understanding and data preparation stage,I hope to construct a customer classification model,which is consist of analysis variables and descriptive variables,and to find a relatively perfect combination in the current state of science and enterprise data to put into the application of enterprise data.At last,I will write a set of specific K-Means algorithm tool,and practice it in the actual production data on the enterprise and analyze and evaluate of the results in detail.Through the research of this issue,that it can meet the practical needs of telecommunications customers and their own business needs,and find every lost user as accurate as possible,and explore all kinds of potential customers before they find their own needs.Operators can find and stimulate these potential customers.