Abstract
We describe the system built by the Documents and Linguistic Technology (DLT) Group at University of Limerick for participation in the French-English Question Answering Task of the Cross Language Evaluation Forum (CLEF). The starting point was the system we used for the same task last year, Besides incremental improvements to the query type identification and named entity recognition components, the query analysis and translation stage was much more sophisticated than last year. This resulted in improved performance.
Original language | English |
---|---|
Pages (from-to) | 404-410 |
Number of pages | 7 |
Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Volume | 3491 |
DOIs | |
Publication status | Published - 2005 |
Event | 5th Workshop of the Cross-Language Evaluation Forum, CLEF 2004: Multilingual Information Access for Text, Speech and Images - Bath, United Kingdom Duration: 15 Sep 2004 → 17 Sep 2004 |