Tracking frequent traversal areas in a Web site via log analysis

In this paper, we explore the problem of analyzing frequently traversed spots in a Web site. We first construct a support knowledge base from Web logs and formulate the problem of tracking frequent traversal areas as a problem of finding strongly connected components in a digraph. We successfully incorporated the tracking features into our test Web site. Our experiments show that our approach can be useful in the innovative applications such as Web promotion, banner ads, and Web content design in the electronic commerce