To Construct Implicit Link Structure by Using Frequent Sequence Miner (FS-Miner)
May Thu Aung, Khin Nwe Ni Tun · 2009
Generally, link analysis is based re-ranking Web-page retrieval in current Web search engine. The primary reason lies in the difference of link structures and data scale between global Web and Web site. In this paper, we propose a framework to construct implicit link structure by using frequent sequence miner (FS-Miner) in the user Web log. And then, the page rank algorithm is applied to re-ranked Web Pages. The popular Web paths are used to extract in userpsilas navigation Web log and predict current userpsilas next step. Our system can solve the sparsity of Web links and the diversity problem of userpsilas access behavior. Moreover, our approach discovers userspsila relevant documents to improve the performance of search results.