The Research On Telecom Customer Churn Prediction System Based On Data Mining

With the rapid development of the economy,the market structure of the telecommunications industry has also undergone enormous changes.The competition among major operators has also become increasingly fierce.The saturation of the market has forced enterprises to gradually shift their focus on “product-centered” to “customer-centered”.Therefore,the important position of the customer in the operator is obvious.On the one hand,various corporate operators are taking the direction of attracting new customers.On the other hand,it is particularly important to find out the customers who have a tendency to lose,and to make retention measures to ensure the stability of existing customers.The booming era of big data makes data mining technology widely used.Of course,this also involves the telecommunications industry.With the continuous development of the telecommunications industry,customer information data has also shown an explosive growth in recent years.How to get valuable information from this large number of seemingly unproductive daily data and to make use of it.For telecom operators,it is no doubt a big chip in the competition.Therefore,data mining technology is a powerful weapon for the telecommunication industry.Through the data mining technology to clean,transform,process,analyze and predict by machine learning model algorithm,it can not only have a clearer understanding of the historical data,but also point out the direction for the future enterprise development.So,this topic adopts a telecom operator customer behavior data,using data mining technology with the scikit-learn toolkit,normative data processing,visualization and modeling analysis with Ada Boost algorithm which is a kind of Emsemble Learning to predict customer behavior,factor analysis,customer churn,and provide decision support.Finally,using Python and Django technology architecture,we build a telecom customer churn prediction system based on B/S structure,and show the whole process of data mining and loss prediction more intuitively.This topic of the telecom customer data analysis and forecasting,bring some guidance for telecom operators,can bring considerable economic benefits for the enterprise,the model and system of scientific and effective,and suitable for practical application,it has an important significance in the field of customer relationship management.