Learning of graph rules for question answering

Diego Mollá; Menno van Zaanen

Learning of graph rules for question answering

Diego Mollá, Menno van Zaanen

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

30 Downloads (Pure)

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 language	English
Title of host publication	Australasian Language Technology Workshop 2005
Subtitle of host publication	proceedings of the workshop, 10-11 December 2005, University of Sydney, Sydney Australia
Place of Publication	Melbourne
Publisher	Australasian Language Technology Association
Pages	15-23
Number of pages	9
ISBN (Print)	097516872X
Publication status	Published - 2005
Event	Australasian Language Technology Workshop (3rd : 2005) - Sydney Duration: 10 Dec 2005 → 11 Dec 2005

Publication series

Name	Proceedings of the Australasian Language Technology Workshop
Publisher	Australasian Language Technology Association
ISSN (Print)	1834-7037

Workshop

Workshop	Australasian Language Technology Workshop (3rd : 2005)
City	Sydney
Period	10/12/05 → 11/12/05

Access to Document

Mq-11883-Author final versionAccepted author manuscript, 252 KBLicence: Unspecified

Cite this

@inproceedings{52b691b328544e82909e94d48e85077a,

title = "Learning of graph rules for question answering",

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.",

author = "Diego Moll{\'a} and {van Zaanen}, Menno",

year = "2005",

language = "English",

isbn = "097516872X",

series = "Proceedings of the Australasian Language Technology Workshop",

publisher = "Australasian Language Technology Association",

pages = "15--23",

booktitle = "Australasian Language Technology Workshop 2005",

note = "Australasian Language Technology Workshop (3rd : 2005) ; Conference date: 10-12-2005 Through 11-12-2005",

}

Mollá, D & van Zaanen, M 2005, Learning of graph rules for question answering. in Australasian Language Technology Workshop 2005: proceedings of the workshop, 10-11 December 2005, University of Sydney, Sydney Australia. Proceedings of the Australasian Language Technology Workshop, Australasian Language Technology Association, Melbourne, pp. 15-23, Australasian Language Technology Workshop (3rd : 2005), Sydney, 10/12/05.

Learning of graph rules for question answering. / Mollá, Diego; van Zaanen, Menno.
Australasian Language Technology Workshop 2005: proceedings of the workshop, 10-11 December 2005, University of Sydney, Sydney Australia. Melbourne: Australasian Language Technology Association, 2005. p. 15-23 (Proceedings of the Australasian Language Technology Workshop).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Learning of graph rules for question answering

AU - Mollá, Diego

AU - van Zaanen, Menno

PY - 2005

Y1 - 2005

N2 - 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.

AB - 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.

M3 - Conference proceeding contribution

SN - 097516872X

T3 - Proceedings of the Australasian Language Technology Workshop

SP - 15

EP - 23

BT - Australasian Language Technology Workshop 2005

PB - Australasian Language Technology Association

CY - Melbourne

T2 - Australasian Language Technology Workshop (3rd : 2005)

Y2 - 10 December 2005 through 11 December 2005

ER -