Investigating Embedded Question Reuse in Question Answering

Abstract

The investigation presented in this paper is a novel method in question answering (QA) that enables a QA system to gain performance through reuse of information in the answer to one question to answer another related question. Our analysis shows that a pair of question in a general open domain QA can have embedding relation through their mentions of noun phrase expressions. We present methods for recognition of embedding property between a pair of questions by focusing on the techniques applied to match noun phrase mentions in the questions. We then take advantage from the discovery of embedding relationship and extract referent named entities corresponding to the noun phrase expression that are present in the answer of one question. The named entities will then be used as significant terms in the query generation phase of the QA system to retrieve more pertinent answers. Finally we discuss on data set resources and system evaluation.