Summary: Named Entity Recognition in Question
Answering of Speech Data
Diego Moll´a Menno van Zaanen Steve Cassidy
Centre for Language Technology, Macquarie University, Sydney, Australia
Our contribution is centred on a study of Named Entity (NE) recognition
on speech transcripts and how it impacts on the accuracy of the final ques-
tion answering system. AnswerFinder was adapted to the task of question
answering on speech transcripts and participated in the QAst pilot track
of the CLEF competition. We have ported AFNER, the NE recogniser of
AnswerFinder, to the set of answer types expected in the QAst track.
AnswerFinder is a question answering system that focuses on shal-
low semantic representations of questions and text [1, 4]. The un-
derlying idea of AnswerFinder is the use of semantic representa-
tions to reduce the impact of paraphrases. The system finds exact
answers to questions in large document collections.