CLIR Experiments at Maryland for TREC-2002: Evidence Combination for Arabic-English Retrieval
Abstract
The focus of the experiments reported in this paper was techniques for combining evidence for crosslanguage retrieval, searching Arabic documents using English queries. Evidence from multiple sources of translation knowledge was combined to estimate translation probabilities, and four techniques for estimating query-language term weights from document-language evidence were tried. A new technique that exploits translation probability information was found to outperform a comparable technique in which that information was not used. Comparative results for three variants of Arabic light stemming are also presented. A simple variant of an existing stemming algorithm was found to result in significantly better retrieval effectiveness.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2002
- Accession Number
- ADA457575
Entities
People
- Douglas W. Oard
- Kareem Darwish
Organizations
- University of Maryland