Using Semantic Constraints for Question Answering

In this paper, we discuss a technique based on semantic constraints to improve the performance and portability of a reformulation-based question answering system. First, we present a method for acquiring semantic-based reformulations automatically. The goal is to generate patterns from correlative articles based on lexical, syntactic and semantic constraints, and a method to evaluate and re-rank candidate answers that satisfy these constraints is adopted. The evaluation on questions from TREC QA tracks 2003 and 2004 shows that the automatically acquired semantic patterns allows us to avoid the manual work of formulating semantically equivalent reformulations, while still reach an acceptable performance.