SemQuest: University of Houston's Semantics-based Question Answering System.
Araly Barrera, Rakesh M Verma, Ryan Vincent · Theory and applications of categories · 2011
This work presents, SemQuest, a questionanswering system used in the TAC 2011 guided summarization task based on semantics and extensions of a previous-developed single-document extractor. Our overall methodology includes: a data cleaning step, linguistic preprocessing among category articles, and a sentence extraction phase. A maximal marginal relevance technique, proposed by Carbonell et al., is also used in SemQuest to reduce redundancy and save space to answer as many category aspects as possible within a short summary extract.