High performance question/answering
- 1 September 2001
- proceedings article
- Published by Association for Computing Machinery (ACM)
- p. 366-374
- https://doi.org/10.1145/383952.384025
Abstract
In this paper we present the features of a Question/Answering (Q/A) system that had unparalleled performance in the TREC-9 evaluations. We explain the accuracy of our system through the unique characteristics of its architecture: (1) usage of a wide-coverage answer type taxonomy; (2) repeated passage retrieval; (3) lexico-semantic feedback loops; (4) extraction of the answers based on machine learning techniques; and (5) answer caching. Experimental results show the effects of each feature on the overall performance of the Q/A system and lead to general conclusions about Q/A from large text collections.Keywords
This publication has 5 references indexed in Scilit:
- The role of lexico-semantic feedback in open-domain textual question-answeringPublished by Association for Computational Linguistics (ACL) ,2001
- Answer extractionPublished by Association for Computational Linguistics (ACL) ,2000
- Information Extraction Supported Question AnsweringPublished by Defense Technical Information Center (DTIC) ,1999
- A new statistical parser based on bigram lexical dependenciesPublished by Association for Computational Linguistics (ACL) ,1996
- WordNetCommunications of the ACM, 1995