Improvement in TF-IDF scheme for Web pages based on the contents of their hyperlinked neighboring pages
Kazunari Sugiyama, Kenji Hatano, Masatoshi Yoshikawa, Shunsuke Uemura · Systems and Computers in Japan · 2005
The TF-IDF scheme is widely used to characterize documents in an information retrieval (IR) system based on the vector space model. However, for documents having a hyperlink structure such as Web pages, the Web page contents can be characterized more accurately by using the contents of hyperlinked neighboring pages. Therefore, in this paper, we propose several techniques for using the contents of hyperlinked neighboring pages to improve the TF-IDF scheme for Web pages and then verify the effectiveness of our techniques. © 2005 Wiley Periodicals, Inc. Syst Comp Jpn, 36(14): 56–68, 2005; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.20189