Analysis Of D-N Hybrid Algorithm Based On China Mobile Customer Churn

The major research of this paper is the problem of customer churn based on datamining technology in the telecommunication industry. The key issue is how to make gooduse of a large number of data through data mining technology to build customer churnmodel. Namely, according to the consumption behaviors and natures of customers who arelost or not lost, through data mining technology to analyze and build the customer churnprediction model. In the process of the analysis, getting the information of customers whohave the greatest probability to lose, then according to the loss of customer’s behaviorsand other related factors to provide decision support for market.In this paper, based on a mobile branch’s customer data, a theoretical research andempirical study method is used and builds D-N hybrid model. During the data researchand build, it will show the detailed explanation of the whole process such as attributeschoosing, data preparation, construction of the model and model evaluation andapplication. In this paper, a more reasonable evaluation method-numerical indicator andthe graphic indicators are used to evaluate the result of the model. The result indicates thatthe hybrid model has better accuracy and hit rates. Meanwhile, the D-N model presentsbetter results than the existing method used by this company. Then using the results of theD-N model, this paper analyses the probability of the mobile branch’s customer churn inthe next month, and the churning customer’s characteristics, such as average fee, length ofservice, gender, the number of calls and so on, and sum up the reason for the loss ofcustomers, and give the corresponding measures to retain customers.Finally in this article I summarized the research work and proposed the content of thefuture research direction and ideas.