The Application Of Data Mining Technology In The Accurate Marketing Of Telecom Enterprises

With the development and progress of the times, the customers’ requirements andexpectations to operators are increasing. Traditional telecommunications mininganalysis only attributes for the users individual, did not consider the relationshipbetween individuals. Monthly consumption indicator of consumers may change whenthe individual’s needs, interests, or economic fluctuations changes, but the relationshipbetween people is stable in the long-term period of time. Analysis based on thecommon features of users in the same groups will be more reliable.The theory of Six Degrees is applied in telecommunication social network,degrees are defined for a single user, each pair of users, groups of users and networkusers. Each degree can be extended to the study of the social networking technology,and make a contribution to the subsequent analysis.The concepts of propinquity, core components and core vesting value are defined,the groups detection algorithm with and without weight based on propinquity isproposed in this article. Customers are divided using the groups detection algorithmwithout weight and the clustering results are evaluated. In order to assist marketing,clustering algorithm is used to analyze the common feature of groups. Logisticregression algorithm is appropriate to measure the weight of family relationships.Potential family groups are forecasted and the results of the model are tested. Thegeometric mean algorithm quantifies the weight of the relationship between usersfrom different operators. To high-value customers who have the tendency of valuechurn, measures must be made to persuade him or her to stay. Also users from operators of competitors who have high values can be urged to transfer operators. Theresearch results can be used to assist in marketing efforts.