Abstract

Previous comparisons of document and query translation suffered difficulty due to differing quality of machine translation in these two opposite directions. We avoid this difficulty by training identical statistical translation models for both translation di- rections using the same training data. We investigate information retrieval between English and French, incorporating both trans- lations directions into both document trans- lation and query translation-based information retrieval, as well as into hybrid systems. We find that hybrids o document and query translation-based systems outperform query translation systems, even human-quality query translation systems. I

Citations

892	Mathematics of Statistical Machine Translation: Parameter Estimation - Brown, Pietra, et al. - 1993
370	Okapi at trec-3 - Robertson, Walker, et al. - 1994
143	Phrasal Translation and Query Expansion Techniques for Cross-Language Information Retrieval - Ballesteros, Croft - 1997
143	Resolving Ambiguity for Cross-Language Retrieval - Ballesteros, Croft - 1998
59	Translingual Information Retrieval: A Comparative Evaluation - Carbonnell, Yang, et al. - 1997
57	Automatic cross-language retrieval using latent semantic indexing - Dumais - 1997
49	Using Clustering and Super Concepts within SMART: TREC 6 - Buckley, Mitra, et al. - 1998
42	A comparative study of query and document translation for cross-language information retrieval - Oard - 1998
24	Document Translation for Cross-Language Text Retrieval at the University of Maryland - Oard, Hackett - 1998
9	Trec-6 ad-hoc retrieval - Franz, Roukos - 1998
4	Fast document translation for cross-language information retrieval - McCarley, Roukos - 1998
3	TREC-5 Ad Hoc Retrieval Using K NearestNeighbors Re-Scoring - Chan, Roukos - 1996