Abstract

QA-by-Dossier-with-Constraints is a new approach to Question Answering whereby candidate answers ’ confidences are adjusted by asking auxiliary questions whose answers constrain the original answers. These constraints emerge naturally from the domain of interest, and enable application of real-world knowledge to QA. We show that our approach significantly improves system performance (75 % relative improvement in F-measure on select question types) and can create a “dossier ” of information about the subject matter in the original question.

Citations

575	CYC: a large-scale investment in knowledge infrastructure - Lenat - 1995
108	Developing a natural language interface to complex data - Hendrix, Sacrdoti, et al. - 1978
83	Overview of the trec 2002 question answering track - Voorhees - 2002
68	An efficient and easily adaptable system for interpreting natural language queries in Computational Linguistics - Warren, Pereira - 1982
57	Procedures as a Representation for Data in a Computer Program for Understanding Natural Language - Winograd - 1971
51	D.: Question-answering by predictive annotation - Prager, Brown, et al. - 2000
48	Logic form transformation of WordNet and its applicability to question answering - Moldovan, Rus - 2001
34	Question answering by passage selection (multitext experiments for TREC-9 - Clarke, Cormack, et al. - 2000
25	A Multi-Strategy and Multi-Source Approach to Question Answering - Chu-Carroll, Prager, et al. - 2003
1	A MultiAgent Approach to using Redundancy and Reinforcement in Question Answering - Prager, Chu-Carroll, et al.