INAOE's Participation at QA@CLEF 2007

This paper describes the system developed by the Language Technologies Lab of INAOE for the Spanish Question Answering task at CLEF 2007. The presented system is centered in a full datadriven architecture that uses information retrieval and machine learning techniques to identify the most probable answers to definition and factoid questions respectively. The major quality of our system is that it mainly relies on the use of lexical information and avoids applying any complex language processing resource such as POS taggers, named entity classifiers, parsers or ontologies. Experimental results indicate that our approach is very effective for answering definition questions from Wikipedia. In contrast, they also reveal that it is very difficult to respond factual questions from this resource solely based on the use of lexical overlaps and redundancy.