为了提高邮政金融网点的营销能力,为营销经理提供精准营销的依据,本文采用数据挖掘的方法针对金融客户的购买行为进行分析,找出购买各类金融产品的客户群体特征。本文对广东邮政某金融网点的客户数据进行了采集并建立起统一视图,然后采用K-means聚类算法对客户数据进行聚类分析,所选用聚类属性是按照当前客户的各类金融产品的百分比进行分析,并且将算法的结果结合客户的基本属性对所有金融客户进行细分,并针对某些共性的客户进行精准的产品推荐。 In order to improve marketing ability of postal financial outlets and provide marketing manager with accurate marketing basis, this paper uses data mining method to analyze the buying behavior of financial customers and find out the characteristics of customer groups who buy various types of financial products. This paper collects and establishes a unified view of the customer data of a financial network in Guangdong Post, and then uses K-means clustering algorithm to cluster the customer data. The selected clustering attributes are based on the current customer’s various financial products Percentage, and the results of the algorithm combined with the basic attributes of customers for all financial customers breakdown, and for some common customers for accurate product recommendations.