Broadband dial-up user behavior identification and analysis

With the evolvement of the Internet and network services, broadband user behavior tends to diversify. It is a rising challenge for the network operators to understand network user behavior. In this paper, we present a general approach for identification and profiling user behavior model by using data mining technique to analyze user activity data. We present analysis of a data set comprised of the dial-up users' usage data in a typical metropolitan area network of a national broadband operator in China. The data recorded activity of over 250,000 unique user accounts. We identify several main behavior groups by using K-means algorithm, and further summarize behavior profile of each user group as main user behavior models. The results show that our approach can properly identify user behavior models. Our results provide a basis for network operators to understand dial-up user behavior, optimize network planning and adjust customer related policies accordingly.