Indexing on semantic roles for question answering

Luiz Augusto Pizzato; Diego Mollá

Indexing on semantic roles for question answering

Luiz Augusto Pizzato, Diego Mollá

School of Computing

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

99 Downloads (Pure)

Abstract

Semantic Role Labeling (SRL) has been used successfully in several stages of automated Question Answering (QA) systems but its inherent slow procedures make it difficult to use at the indexing stage of the document retrieval component. In this paper we confirm the intuition that SRL at indexing stage improves the performance of QA and propose a simplified technique named the Question Prediction Language Model (QPLM), which provides similar information with a much lower cost. The methods were tested on four different QA systems and the results suggest that QPLM can be used as a good compromise between speed and accuracy.

Original language	English
Title of host publication	IRQA 2008
Subtitle of host publication	Proceedings of the 2nd Workshop on Information Retrieval for Question Answering, August 24, 2008, Manchester, UK
Editors	Mark A. Greenwood
Place of Publication	Morristown, N.J.
Publisher	Association for Computational Linguistics
Pages	74-81
Number of pages	8
ISBN (Print)	9781905593552
Publication status	Published - 2008
Event	Coling 2008 Workshop on Information Retrieval for Question Answering (2nd : 2008) - Manachester Duration: 18 Aug 2008 → 22 Aug 2008

Workshop

Workshop	Coling 2008 Workshop on Information Retrieval for Question Answering (2nd : 2008)
City	Manachester
Period	18/08/08 → 22/08/08

Bibliographical note

Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Access to Document

Mq-11895-Publisher version (open access)Final published version, 120 KBLicence: Unspecified

Cite this

@inproceedings{02e1e7d3c3874571a3a33ffdbe7da605,

title = "Indexing on semantic roles for question answering",

abstract = "Semantic Role Labeling (SRL) has been used successfully in several stages of automated Question Answering (QA) systems but its inherent slow procedures make it difficult to use at the indexing stage of the document retrieval component. In this paper we confirm the intuition that SRL at indexing stage improves the performance of QA and propose a simplified technique named the Question Prediction Language Model (QPLM), which provides similar information with a much lower cost. The methods were tested on four different QA systems and the results suggest that QPLM can be used as a good compromise between speed and accuracy.",

author = "Pizzato, {Luiz Augusto} and Diego Moll{\'a}",

note = "Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.; Coling 2008 Workshop on Information Retrieval for Question Answering (2nd : 2008) ; Conference date: 18-08-2008 Through 22-08-2008",

year = "2008",

language = "English",

isbn = "9781905593552",

pages = "74--81",

editor = "Greenwood, {Mark A.}",

booktitle = "IRQA 2008",

publisher = "Association for Computational Linguistics",

}

Pizzato, LA & Mollá, D 2008, Indexing on semantic roles for question answering. in MA Greenwood (ed.), IRQA 2008: Proceedings of the 2nd Workshop on Information Retrieval for Question Answering, August 24, 2008, Manchester, UK. Association for Computational Linguistics, Morristown, N.J., pp. 74-81, Coling 2008 Workshop on Information Retrieval for Question Answering (2nd : 2008), Manachester, 18/08/08.

Indexing on semantic roles for question answering. / Pizzato, Luiz Augusto; Mollá, Diego.
IRQA 2008: Proceedings of the 2nd Workshop on Information Retrieval for Question Answering, August 24, 2008, Manchester, UK. ed. / Mark A. Greenwood. Morristown, N.J.: Association for Computational Linguistics, 2008. p. 74-81.

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

TY - GEN

T1 - Indexing on semantic roles for question answering

AU - Pizzato, Luiz Augusto

AU - Mollá, Diego

N1 - Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2008

Y1 - 2008

N2 - Semantic Role Labeling (SRL) has been used successfully in several stages of automated Question Answering (QA) systems but its inherent slow procedures make it difficult to use at the indexing stage of the document retrieval component. In this paper we confirm the intuition that SRL at indexing stage improves the performance of QA and propose a simplified technique named the Question Prediction Language Model (QPLM), which provides similar information with a much lower cost. The methods were tested on four different QA systems and the results suggest that QPLM can be used as a good compromise between speed and accuracy.

AB - Semantic Role Labeling (SRL) has been used successfully in several stages of automated Question Answering (QA) systems but its inherent slow procedures make it difficult to use at the indexing stage of the document retrieval component. In this paper we confirm the intuition that SRL at indexing stage improves the performance of QA and propose a simplified technique named the Question Prediction Language Model (QPLM), which provides similar information with a much lower cost. The methods were tested on four different QA systems and the results suggest that QPLM can be used as a good compromise between speed and accuracy.

M3 - Conference proceeding contribution

SN - 9781905593552

SP - 74

EP - 81

BT - IRQA 2008

A2 - Greenwood, Mark A.

PB - Association for Computational Linguistics

CY - Morristown, N.J.

T2 - Coling 2008 Workshop on Information Retrieval for Question Answering (2nd : 2008)

Y2 - 18 August 2008 through 22 August 2008

ER -

Indexing on semantic roles for question answering

Abstract

Workshop

Bibliographical note

Access to Document

Fingerprint

Cite this