Customer Behavior Clustering Using SVM

摘要：

In order to supply better service for network customers, deeply analyzing customers behavior is required. This paper extracts three features from customers network behavior which is divided into different categories, such as browsing news, downloading shared resources and real-time communications etc. Support vector machine is used to perform clustering, thanks to its fast and valid executing, especially in the situation of small datasets. Using the analysis results, we can make our applications and services more personalized and easier to be used.

关键词： SVM Customer behavior analysis Clustering

作者: Zhongying Yang Xiaolong Su

作者单位: Dept of Compute Science & Technology China University of Mining & Technology Xuzhou Jiangsu 221000, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 49-52

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）