Semantic-head-driven generation
Computational Linguistics
The anatomy of a large-scale hypertextual Web search engine
WWW7 Proceedings of the seventh international conference on World Wide Web 7
Authoritative sources in a hyperlinked environment
Journal of the ACM (JACM)
Head-driven statistical models for natural language parsing
Head-driven statistical models for natural language parsing
COLING '98 Proceedings of the 17th international conference on Computational linguistics - Volume 1
Parameter estimation for probabilistic finite-state transducers
ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Finding similar questions in large question and answer archives
Proceedings of the 14th ACM international conference on Information and knowledge management
Learning structured prediction models: a large margin approach
ICML '05 Proceedings of the 22nd international conference on Machine learning
The Proposition Bank: An Annotated Corpus of Semantic Roles
Computational Linguistics
Respect my authority!: HITS without hyperlinks, utilizing cluster-based language models
SIGIR '06 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval
Learning question classifiers: the role of semantic information
Natural Language Engineering
Experiments with interactive question-answering
ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics
Methods for using textual entailment in open-domain question answering
ACL-44 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
Learning to recognize features of valid textual entailments
HLT-NAACL '06 Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics
The third PASCAL recognizing textual entailment challenge
RTE '07 Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing
Dependency-based paraphrasing for recognizing textual entailment
RTE '07 Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing
A discourse commitment-based framework for recognizing textual entailment
RTE '07 Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing
| Hi-index | 0.00 |
This paper presents a novel approach to textual question answering (QA) which identifies answers to natural language questions by leveraging large collections of question-answer pairs extracted from web sources or generated automatically from text. Instead of using the traditional retrieve-and-rerank approach common to most previous approaches to factoid question answering, we introduce a new model of answer authority which allows question-answering systems to estimate the quality of answers not just in isolation - but in the larger context of the information contained in the corpus as a whole. Our approach in this paper hinges on the creation of a new representation of the information stored in a document collection, known as a Question-Answer Database (QUAB).We assume that a QUAB represents a weighted directed graph consisting of the set of factoid question-answer pairs (QAP) that can be asked - and answered - given the content of a corpus. Once a set of QAP have been generated, we use inferential relationships identified by a system for recognizing textual entailment in order to construct the link structure necessary to compute the authority of each interconnected question or answer. We have found access to the QUAB graph not only improves the accuracy of current answer retrieval techniques, but also allows for the determination of when no valid answer can be found for a question in a text corpus. Experimental results show that the authority derived from a graph of question-answer pairs can increase the performance of a factoid QA system by nearly 30%.