AnswerFinder at QAst 2007: Named Entity recognition for QA on speech transcripts

Diego Mollá, Steve Cassidy, Menno Van Zaanen

Research output: Contribution to journalConference paperResearch

Abstract

Macquarie University's contribution to the QAst track of CLEF is centered on a study of Named Entity (NE) recognition on speech transcripts, and how such NE recognition impacts on the accuracy of the final question answering system. We have ported AFNER, the NE recogniser of the AnswerFinder question-answering project, to the types of answer types expected in the QAst track. AFNER uses a combination of regular expressions, lists of names (gazetteers) and machine learning. The machine learning component is a Maximum Entropy classifier and was trained on a development set of the AMI corpus. Problems with scalability of the system and errors of the extracted annotation lead to relatively poor performance in general, though the system was second (out of three participants) in one of the QAst subtasks.

LanguageEnglish
Pages1-9
Number of pages9
JournalCEUR Workshop Proceedings
Volume1173
Publication statusPublished - 2007

Fingerprint

Learning systems
Scalability
Classifiers
Entropy

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.

Cite this

@article{babb1d29d28a4378a97376848ef03573,
title = "AnswerFinder at QAst 2007: Named Entity recognition for QA on speech transcripts",
abstract = "Macquarie University's contribution to the QAst track of CLEF is centered on a study of Named Entity (NE) recognition on speech transcripts, and how such NE recognition impacts on the accuracy of the final question answering system. We have ported AFNER, the NE recogniser of the AnswerFinder question-answering project, to the types of answer types expected in the QAst track. AFNER uses a combination of regular expressions, lists of names (gazetteers) and machine learning. The machine learning component is a Maximum Entropy classifier and was trained on a development set of the AMI corpus. Problems with scalability of the system and errors of the extracted annotation lead to relatively poor performance in general, though the system was second (out of three participants) in one of the QAst subtasks.",
author = "Diego Moll{\'a} and Steve Cassidy and {Van Zaanen}, Menno",
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.",
year = "2007",
language = "English",
volume = "1173",
pages = "1--9",
journal = "CEUR Workshop Proceedings",
issn = "1613-0073",
publisher = "RWTH Aachen University",

}

AnswerFinder at QAst 2007 : Named Entity recognition for QA on speech transcripts. / Mollá, Diego; Cassidy, Steve; Van Zaanen, Menno.

In: CEUR Workshop Proceedings, Vol. 1173, 2007, p. 1-9.

Research output: Contribution to journalConference paperResearch

TY - JOUR

T1 - AnswerFinder at QAst 2007

T2 - CEUR Workshop Proceedings

AU - Mollá,Diego

AU - Cassidy,Steve

AU - Van Zaanen,Menno

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 - 2007

Y1 - 2007

N2 - Macquarie University's contribution to the QAst track of CLEF is centered on a study of Named Entity (NE) recognition on speech transcripts, and how such NE recognition impacts on the accuracy of the final question answering system. We have ported AFNER, the NE recogniser of the AnswerFinder question-answering project, to the types of answer types expected in the QAst track. AFNER uses a combination of regular expressions, lists of names (gazetteers) and machine learning. The machine learning component is a Maximum Entropy classifier and was trained on a development set of the AMI corpus. Problems with scalability of the system and errors of the extracted annotation lead to relatively poor performance in general, though the system was second (out of three participants) in one of the QAst subtasks.

AB - Macquarie University's contribution to the QAst track of CLEF is centered on a study of Named Entity (NE) recognition on speech transcripts, and how such NE recognition impacts on the accuracy of the final question answering system. We have ported AFNER, the NE recogniser of the AnswerFinder question-answering project, to the types of answer types expected in the QAst track. AFNER uses a combination of regular expressions, lists of names (gazetteers) and machine learning. The machine learning component is a Maximum Entropy classifier and was trained on a development set of the AMI corpus. Problems with scalability of the system and errors of the extracted annotation lead to relatively poor performance in general, though the system was second (out of three participants) in one of the QAst subtasks.

UR - http://www.scopus.com/inward/record.url?scp=84921966975&partnerID=8YFLogxK

M3 - Conference paper

VL - 1173

SP - 1

EP - 9

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

ER -