TCS-TR-A-07-24

Date: Tue Feb 13 15:21:54 2007

Title: Flexible Wrappers for Keyword-Related Information

Authors: Atsuyoshi Nakamura, Hiroyuki Hasegawa, Tomoya Saito, Mineichi Kudo

Contact:

First name: Atsuyoshi
Last name: Nakamura
Address: Graduate School of Information Science and Technology, Hokkaido University Kita 14, Nishi 9, Kita-ku, Sapporo, 060-0814 Japan
Email: atsu@main.ist.hokudai.ac.jp

Abstract. We propose an approach for realizing flexible wrappers, the keyword-related target item extraction based on structure and contents. Our approach composed of four phases, the (flexible) structure-based extraction phase, the content-based filtering phase, the decision phase and the error correction phase. We show one implementation of our approach and report encouraging results on our preliminary experiments that indicate practical usefulness of our method for information exploration.