Minimum Bayes Risk based Answer Re-ranking for Question Answering
The 51st Annual Meeting of the Association for Computational Linguistics - Short Papers (ACL Short Papers 2013)
Sofia, Bulgaria, August 4-9, 2013
This paper presents two minimum Bayes risk (MBR) based Answer Re-ranking (MBRAR) approaches for the question answering (QA) task. The first approach re-ranks single QA system’s outputs by using a traditional MBR model, which measures correlations between answer candidates; while the second approach re-ranks the combined outputs of multiple QA systems with heterogenous answer extraction components by using a mixture model based MBR model. Evaluations are performed on factoid questions selected from two different domains: Web and Jeopardy!, and significant improvements are achieved on all data sets.
Conference Manager (V2.61.0 - Rev. 2792M)