A tree kernel approach to question and answer classification in question answering systems

Alessandro Moschitti, Roberto Basili

Research output: Contribution to conferencePaper

12 Citations (Scopus)

Abstract

A critical step in Question Answering design is the definition of the models for question focus identification and answer extraction. In case of factoid questions, we can use a question classifier (trained according to a target taxonomy) and a named entity recognizer. Unfortunately, this latter cannot be applied to generate answers related to non-factoid questions. In this paper, we tackle such problem by designing classifiers of non-factoid answers. As the feature design for this learning task is very complex, we take advantage of tree kernels to generate large feature set from the syntactic parse trees of passages relevant to the target question. Such kernels encode syntactic and lexical information in Support Vector Machines which can decide if a sentence focuses on a target taxonomy subject. The experiments with SVMs on the TREC 10 dataset show that our approach is an interesting future research.

Original languageEnglish
Pages1510-1513
Number of pages4
Publication statusPublished - 1 Jan 2006
Event5th International Conference on Language Resources and Evaluation, LREC 2006 - Genoa, Italy
Duration: 22 May 200628 May 2006

Other

Other5th International Conference on Language Resources and Evaluation, LREC 2006
CountryItaly
CityGenoa
Period22/5/0628/5/06

Fingerprint

taxonomy
experiment
learning
Kernel
Question Answering
Syntax
Taxonomy
Classifier
Support Vector Machine
Entity
Experiment

ASJC Scopus subject areas

  • Education
  • Library and Information Sciences
  • Linguistics and Language
  • Language and Linguistics

Cite this

Moschitti, A., & Basili, R. (2006). A tree kernel approach to question and answer classification in question answering systems. 1510-1513. Paper presented at 5th International Conference on Language Resources and Evaluation, LREC 2006, Genoa, Italy.

A tree kernel approach to question and answer classification in question answering systems. / Moschitti, Alessandro; Basili, Roberto.

2006. 1510-1513 Paper presented at 5th International Conference on Language Resources and Evaluation, LREC 2006, Genoa, Italy.

Research output: Contribution to conferencePaper

Moschitti, A & Basili, R 2006, 'A tree kernel approach to question and answer classification in question answering systems' Paper presented at 5th International Conference on Language Resources and Evaluation, LREC 2006, Genoa, Italy, 22/5/06 - 28/5/06, pp. 1510-1513.
Moschitti A, Basili R. A tree kernel approach to question and answer classification in question answering systems. 2006. Paper presented at 5th International Conference on Language Resources and Evaluation, LREC 2006, Genoa, Italy.
Moschitti, Alessandro ; Basili, Roberto. / A tree kernel approach to question and answer classification in question answering systems. Paper presented at 5th International Conference on Language Resources and Evaluation, LREC 2006, Genoa, Italy.4 p.
@conference{01a0d46d1b9a442f893a29feb90615ee,
title = "A tree kernel approach to question and answer classification in question answering systems",
abstract = "A critical step in Question Answering design is the definition of the models for question focus identification and answer extraction. In case of factoid questions, we can use a question classifier (trained according to a target taxonomy) and a named entity recognizer. Unfortunately, this latter cannot be applied to generate answers related to non-factoid questions. In this paper, we tackle such problem by designing classifiers of non-factoid answers. As the feature design for this learning task is very complex, we take advantage of tree kernels to generate large feature set from the syntactic parse trees of passages relevant to the target question. Such kernels encode syntactic and lexical information in Support Vector Machines which can decide if a sentence focuses on a target taxonomy subject. The experiments with SVMs on the TREC 10 dataset show that our approach is an interesting future research.",
author = "Alessandro Moschitti and Roberto Basili",
year = "2006",
month = "1",
day = "1",
language = "English",
pages = "1510--1513",
note = "5th International Conference on Language Resources and Evaluation, LREC 2006 ; Conference date: 22-05-2006 Through 28-05-2006",

}

TY - CONF

T1 - A tree kernel approach to question and answer classification in question answering systems

AU - Moschitti, Alessandro

AU - Basili, Roberto

PY - 2006/1/1

Y1 - 2006/1/1

N2 - A critical step in Question Answering design is the definition of the models for question focus identification and answer extraction. In case of factoid questions, we can use a question classifier (trained according to a target taxonomy) and a named entity recognizer. Unfortunately, this latter cannot be applied to generate answers related to non-factoid questions. In this paper, we tackle such problem by designing classifiers of non-factoid answers. As the feature design for this learning task is very complex, we take advantage of tree kernels to generate large feature set from the syntactic parse trees of passages relevant to the target question. Such kernels encode syntactic and lexical information in Support Vector Machines which can decide if a sentence focuses on a target taxonomy subject. The experiments with SVMs on the TREC 10 dataset show that our approach is an interesting future research.

AB - A critical step in Question Answering design is the definition of the models for question focus identification and answer extraction. In case of factoid questions, we can use a question classifier (trained according to a target taxonomy) and a named entity recognizer. Unfortunately, this latter cannot be applied to generate answers related to non-factoid questions. In this paper, we tackle such problem by designing classifiers of non-factoid answers. As the feature design for this learning task is very complex, we take advantage of tree kernels to generate large feature set from the syntactic parse trees of passages relevant to the target question. Such kernels encode syntactic and lexical information in Support Vector Machines which can decide if a sentence focuses on a target taxonomy subject. The experiments with SVMs on the TREC 10 dataset show that our approach is an interesting future research.

UR - http://www.scopus.com/inward/record.url?scp=84886224960&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886224960&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84886224960

SP - 1510

EP - 1513

ER -