Paper: Topic Indexing and Retrieval for Factoid QA

ACL ID W08-1809
Title Topic Indexing and Retrieval for Factoid QA
Venue Coling 2008: Proceedings of the workshop on Human Judgements in Computational Linguistics
Year 2008

The method of Topic Indexing and Re- trieval for QA persented in this paper enables fast and efficent QA for ques- tions with named entity answers. This is achieved by identifying all possible named entity answers in a corpus off-line and gathering all possible evidence for their di- rect retrieval as answer candidates using standard IR techniques. An evaluation of this method on 377 TREC questions pro- duced a score of 0.342 in Accuracy and 0.413 in Mean Reciprocal Rank (MRR).