An approach for answer selection in Question Answering based on semantic relations
Ana Mendes and Luísa Coheur
A usual strategy to select the final answer in factoid Question-Answering (QA) relies on redundancy. A score is given to each candidate answer as a func- tion of its frequency of occurrence, and the final answer is selected from the set of candidates sorted in decreasing order of score. For that purpose, sys- tems often try to group together semantically equiv- alent answers. However, they hold several other se- mantic relations, such as inclusion, which are not considered, and candidates are mostly seen inde- pendently, as competitors. Our hypothesis is that not just equivalence, but other relations between candidate answers have im- pact on the performance of a redundancy-based QA system. In this paper, we describe experimen- tal studies to back up this hypothesis. Our find- ings show that, with relatively simple techniques to recognize relations, systems’ accuracy can be im- proved for answers of categories NUMBER, DATE and ENTITY.