Customer Segmentation Based on Survival Character

"Customer Segmentation" is an increasingly pressing issue in today's over-competitive commercial arena. More and more literatures have researched the application of data mining technology in customer segmentation, and achieved sound effectives. But most of them segment customer only by single data mining technology from a special point, rather than from a systematical framework. Furthermore, one of the key purposes of customer segmentation is customer retention. Although previous segment methods may identify which group needs more care, it is unable to identify customer churn trend for adopting different action. This paper focus on proposing a customer segmentation framework based on data mining and constructs a new customer segmentation method based on customer survival character. The new customer segmentation method constituted of two steps. Firstly, using K-means clustering arithmetic cluster customer into different segments which has the similar survival function (churn trend) inside. Secondly, the method uses survival analysis to predict each cluster's survival/hazard function to test the validity of clustering and identify customer churn trend. The method mentioned above was applied in a dataset from China Telecommunications Company, which result proposes some useful management measures and suggestions. This paper also suggests some propositions for further research.