Query expansion with terms selected using lexical cohesion analysis of documents |
| |
| Authors: | Olga Vechtomova Murat Karamuftuoglu |
| |
| Institution: | 1. Department of Management Sciences, University of Waterloo, 200 University Avenue West, Waterloo, Ont., Canada N2L 3GE;2. Department of Computer Engineering, Bilkent University, Bilkent, 06800 Ankara, Turkey |
| |
| Abstract: | We present new methods of query expansion using terms that form lexical cohesive links between the contexts of distinct query terms in documents (i.e., words surrounding the query terms in text). The link-forming terms (link-terms) and short snippets of text surrounding them are evaluated in both interactive and automatic query expansion (QE). We explore the effectiveness of snippets in providing context in interactive query expansion, compare query expansion from snippets vs. whole documents, and query expansion following snippet selection vs. full document relevance judgements. The evaluation, conducted on the HARD track data of TREC 2005, suggests that there are considerable advantages in using link-terms and their surrounding short text snippets in QE compared to terms selected from full-texts of documents. |
| |
| Keywords: | Interactive information retrieval Query expansion Relevance feedback Lexical cohesion |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
