Research On Telecom Customer Retention And Loss Based On Data Mining

The arrival of the 4G era,the three major telecom operators in the key management of 4G customers and customers from the past "to save with extension" to "to extension with save",the development of telecom operators is the stock of customers,so the stock of customers and customer value and maintain stock business has a direct impact on the operators of the performance guarantee.Improve the stock of customer loyalty,mining and enhance the value of the stock of customers,is to improve the operator's scale development of the necessary conditions.Telecom operators adapt to the stock of customers management under the4 G era demand,grasp the new stock of customers management in order to better mining and improve the value of the customer.Study on data mining and loss of telecom customer retention is based on massive data on Telecom cumulative mining and forecast the reasons for the loss of customer,in order to predict the reason that retention strategies and methods of all kinds of reasons for the loss of the corresponding customer.The work of this paper is to accumulate in the telecom industry customer basic information,customer information is on the basis of CRISP-DM data mining process is divided into six stages: the publishing business understanding,data understanding,data preparation,modeling,evaluation,model,analysis software SPSS and data mining tool Clementine prediction model is established by statistics.This paper introduces the data mining in the telecom business development background and theoretical knowledge of data mining,and a detailed description of the clustering algorithm and decision tree algorithm,in the process of modeling,the effective data cleaning,conversion of data,finally established by clustering algorithm and decision tree algorithm is evaluated on the index model.In the maintenance of retention work preliminary analysis of the reasons for the loss of customers and assess the value of customers,the reasons for the loss of different retention strategies to reduce retention costs and improve the success rate of preservation.