Paper: Passage Retrieval for Question Answering using Sliding Windows

ACL ID W08-1804
Title Passage Retrieval for Question Answering using Sliding Windows
Venue Coling 2008: Proceedings of the workshop on Human Judgements in Computational Linguistics
Session
Year 2008
Authors

The information retrieval (IR) commu- nity has investigated many different tech- niques to retrieve passages from large col- lections of documents for question answer- ing (QA). In this paper, we specifically ex- amine and quantitatively compare the im- pact of passage retrieval for QA using slid- ing windows and disjoint windows. We consider two different data sets, the TREC 2002–2003 QA data set, and 93 why- questions against INEX Wikipedia. We discovered that, compared to disjoint win- dows, using sliding windows results in im- proved performance of TREC-QA in terms of TDRR, and in improved performance of why-QA in terms of success@n and MRR.