论文信息 - The University of Groningen at QA@CLEF 2006: Using Syntactic Knowledge for QA

The University of Groningen at QA@CLEF 2006: Using Syntactic Knowledge for QA

We describe our system for the monolingual Dutch and multilingual English to Dutch QA tasks. First, we give a brief outline of the architecture of our QA-system, which makes heavy use of syntactic information. Next, we describe the modules that were improved or developed esepcially for the CLEF tasks, i.e. (1) incorporation of syntactic knowledge in the IR-engine, (2) incorporation of lexical equivalences, (3) incorporation of coreference resolution for o-line answer extraction, (4) treatment of temporally restricted questions, (5) treatment of denition questions, and (6) a baseline multilingual (English to Dutch) QA system, which uses a combination of Systran and Wikipedia (for term recognition and translation) for question translation. For non-list questions, 31% of the highest ranked answers returned by the monolingual system were correct and 20% of the answers returned by the multilingual system.

[1] Maarten de Rijke,et al. Overview of the CLEF 2004 Multilingual Question Answering Track , 2004, CLEF.

[2] Jörg Tiedemann,et al. Question Answering for Dutch using Dependency Relations , 2005, CLEF.

[3] Gosse Bouma,et al. Linguistic Knowledge and Question Answering , 2006 .

[4] Jörg Tiedemann. Improving Passage Retrieval in Question Answering Using NLP , 2005, EPIA.

[5] Gosse Bouma. Reasoning over Dependency Relations for QA , 2005 .

[6] Gertjan van Noord,et al. Alpino: Wide-coverage Computational Analysis of Dutch , 2000, CLIN.

[7] Anne-Laure Ligozat,et al. Evaluation and Improvement of Cross-Lingual Question AnsweringStrategies , 2006 .

[8] Gosse Bouma,et al. Learning to Identify Definitions using Syntactic Features , 2006, Learning Structured Information@EACL.