With the development of Data Mining and the growth of information, many enterprises and cooperates begin to use the Data Mining technique to analyze and process the large amount of data which is accumulated in the business activities in order to find out important rules for marketing strategy. Customer clustering is a crucial issue. According to customer clustering, similar customer groups cluster on the basis of their personal and behavioral attributes. Then, the enterprises can work out different marketing strategies and apply them to each customer group respectively. The thesis focuses on two questions about clustering formation. First, several typical clustering algorithms and their complexities are introduced. Meanwhile, the application and efficiency of the clustering algorithm are discussed in it, and then the effectual arithmetic on how to carry through characteristic distill in the clustering formation is given out.