Efficiently computing frequent tree-like topology patterns in a web environment

Data mining recently emerges from a number of real-life applications, such as banking, supermarketing, and interneting. In this paper, we investigate a novel data mining problem derived from the applications of WWW. We will present an efficient algorithm for solving the problem. Our initial implementation suggests that the algorithm is very efficient and scalable in practice.