Study On Clustering Method Of Car Customer From The Perspective Of Cluster Analysis

With the rapid development of economy in China, the living standards have been improved and automotive market in China has been expanded. In 2010, China's auto sale volume was up to 12million, which was writing a history of car development in China once again. As an auto market close to ten million grade, its own characteristics are gradually showing up and its internal pattern is playing a big role. With the continual and effective growth in the automotive market, as the world's second largest auto consuming country, China's auto market has become an important part of global market of independent personality and characteristics. Therefore, understanding and meeting the need of customers, continuously improving the product quality, and improving service level have become important weapons to gain advantages in competitive market.Customer segmentation is a very important business issue. Companies provide personalized customer service and products to different customers. Therefore, in connect with the design of products and service, companies should firstly have an understanding of the customer base. With the personal information, behavioral attributes and consuming information of customers, we could cluster the customers in order to deepen the understanding of customers and develop appropriate marketing strategies.This paper studies the clustering problem of automotive customers. Firstly, this paper studies the relevant research at home and abroad and summarizes the contents and methods of the previous studies. Secondly, this paper introduces the theories of customer clustering problem, such as the conception, principles and features of customer segmentation, and then introduces the commonly used method of customer segmentation and the processing steps. Thirdly, this paper studies the necessarity of automotive customers clustering and introduces the method this paper has used to cluster automotive customers. Fourthly, this paper has proposed the indicators(properties) of this study and according to the data collected, principal component analysis is used to eliminate the correlation among the indicators and then cluster analysis is used to achieve the car customer clustering, At last, this paper concludes the car development rules of clustering customers.