Advanced structural representations for question classification and answer re-ranking

Silvia Quarteroni; Alessandro Moschitti; Suresh Manandhar; Roberto Basili

Advanced structural representations for question classification and answer re-ranking

Silvia Quarteroni, Alessandro Moschitti, Suresh Manandhar, Roberto Basili

Computer Science

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

Abstract

In this paper, we study novel structures to represent information in three vital tasks in question answering: question classification, answer classification and answer reranking. We define a new tree structure called PAS to represent predicate-argument relations, as well as a new kernel function to exploit its representative power. Our experiments with Support Vector Machines and several tree kernel functions suggest that syntactic information helps specific task as question classification, whereas, when data sparseness is higher as in answer classification, studying coarse semantic information like PAS is a promising research area.

Original language	English
Title of host publication	Advances in Information Retrieval
Editors	G Amati, C Carpineto, G Romano
Place of Publication	BERLIN
Publisher	Springer
Pages	234-245
Number of pages	12
ISBN (Print)	978-3-540-71494-1
Publication status	Published - 2007
Event	29th European Conference in Information Retrieval Research (ECIR 2007) - Rome Duration: 2 Apr 2007 → 5 Apr 2007

Conference

Conference	29th European Conference in Information Retrieval Research (ECIR 2007)
City	Rome
Period	2/04/07 → 5/04/07

Cite this

@inproceedings{fd5c98cd76bd4811919a69167b86ceb4,

title = "Advanced structural representations for question classification and answer re-ranking",

abstract = "In this paper, we study novel structures to represent information in three vital tasks in question answering: question classification, answer classification and answer reranking. We define a new tree structure called PAS to represent predicate-argument relations, as well as a new kernel function to exploit its representative power. Our experiments with Support Vector Machines and several tree kernel functions suggest that syntactic information helps specific task as question classification, whereas, when data sparseness is higher as in answer classification, studying coarse semantic information like PAS is a promising research area.",

author = "Silvia Quarteroni and Alessandro Moschitti and Suresh Manandhar and Roberto Basili",

year = "2007",

language = "English",

isbn = "978-3-540-71494-1",

pages = "234--245",

editor = "G Amati and C Carpineto and G Romano",

booktitle = "Advances in Information Retrieval",

publisher = "Springer",

address = "Germany",

note = "29th European Conference in Information Retrieval Research (ECIR 2007) ; Conference date: 02-04-2007 Through 05-04-2007",

}

TY - GEN

T1 - Advanced structural representations for question classification and answer re-ranking

AU - Quarteroni, Silvia

AU - Moschitti, Alessandro

AU - Manandhar, Suresh

AU - Basili, Roberto

PY - 2007

Y1 - 2007

N2 - In this paper, we study novel structures to represent information in three vital tasks in question answering: question classification, answer classification and answer reranking. We define a new tree structure called PAS to represent predicate-argument relations, as well as a new kernel function to exploit its representative power. Our experiments with Support Vector Machines and several tree kernel functions suggest that syntactic information helps specific task as question classification, whereas, when data sparseness is higher as in answer classification, studying coarse semantic information like PAS is a promising research area.

AB - In this paper, we study novel structures to represent information in three vital tasks in question answering: question classification, answer classification and answer reranking. We define a new tree structure called PAS to represent predicate-argument relations, as well as a new kernel function to exploit its representative power. Our experiments with Support Vector Machines and several tree kernel functions suggest that syntactic information helps specific task as question classification, whereas, when data sparseness is higher as in answer classification, studying coarse semantic information like PAS is a promising research area.

M3 - Conference contribution

SN - 978-3-540-71494-1

SP - 234

EP - 245

BT - Advances in Information Retrieval

A2 - Amati, G

A2 - Carpineto, C

A2 - Romano, G

PB - Springer

CY - BERLIN

T2 - 29th European Conference in Information Retrieval Research (ECIR 2007)

Y2 - 2 April 2007 through 5 April 2007

ER -