Improvement in TF-IDF scheme for web pages based on the contents of their hyperlinked neighboring pages

被引：0

作者：

Sugiyama, Kazunari ^{[1
,3
,4
,5
,6
,7
]}

Hatano, Kenji ^{[1
,3
,5
,8
]}

Yoshikawa, Masatoshi ^{[2
,3
,5
,8
,9
]}

Uemura, Shunsuke ^{[1
,3
,5
,7
,10
]}

机构：

[1] Graduate School of Information Science, Nara Institute of Science and Technology, Ikoma, 630-0192, Japan

[2] Information Technology Center, Nagoya University, Nagoya, 464-8601, Japan

[3] Information Processing Society of Japan

[4] Japanese Society for Artificial Intelligence

[5] Association for Computing Machinery

[6] American Association for Artificial Intelligence

[7] IEEE

[8] IEEE Computer Society

[9] Information Technology Center, Nagoya University

[10] Graduate School of Information Science, Nara Institute of Science and Technology

来源：

Systems and Computers in Japan | 2005年 / 36卷 / 14期

关键词：

Information retrieval - Mathematical models - Vectors;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

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 verity the effectiveness of our techniques. © 2005 Wiley Periodicals, Inc.

引用

页码：56 / 68