Learning of graph rules for question answering

Diego Mollá, Menno van Zaanen

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

AnswerFinder is a framework for the development of question-answering systems. AnswerFinder is currently being used to test the applicability of graph representations for the detection and extraction of answers. In this paper we briefly describe AnswerFinder and introduce our method to learn graph patterns that link questions with their corresponding answers in arbitrary sentences. The method is based on the translation of the logical forms of questions and answer sentences into graphs, and the application of operations based on graph overlaps and the construction of paths within graphs. The method is general and can be applied to any graph-based representation of the contents of questions and answers.
Original languageEnglish
Title of host publicationAustralasian Language Technology Workshop 2005
Subtitle of host publicationproceedings of the workshop, 10-11 December 2005, University of Sydney, Sydney Australia
Place of PublicationMelbourne
PublisherAustralasian Language Technology Association
Pages15-23
Number of pages9
ISBN (Print)097516872X
Publication statusPublished - 2005
EventAustralasian Language Technology Workshop (3rd : 2005) - Sydney
Duration: 10 Dec 200511 Dec 2005

Publication series

NameProceedings of the Australasian Language Technology Workshop
PublisherAustralasian Language Technology Association
ISSN (Print)1834-7037

Workshop

WorkshopAustralasian Language Technology Workshop (3rd : 2005)
CitySydney
Period10/12/0511/12/05

Cite this