Encoding semantic resources in syntactic structures for passage reranking

Kateryna Tymoshenko, Alessandro Moschitti, Aliaksei Severyn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Citations (Scopus)

Abstract

In this paper, we propose to use semantic knowledge from Wikipedia and large-scale structured knowledge datasets available as Linked Open Data (LOD) for the answer passage reranking task. We represent question and candidate answer passages with pairs of shallow syntactic/semantic trees, whose constituents are connected using LOD. The trees are processed by SVMs and tree kernels, which can automatically exploit tree fragments. The experiments with our SVM rank algorithm on the TREC Question Answering (QA) corpus show that the added relational information highly improves over the state of the art, e.g., about 15.4% of relative improvement in P@1.

Original languageEnglish
Title of host publication14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014
PublisherAssociation for Computational Linguistics (ACL)
Pages664-672
Number of pages9
ISBN (Print)9781632663962
Publication statusPublished - 1 Jan 2014
Event14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014 - Gothenburg, Sweden
Duration: 26 Apr 201430 Apr 2014

Other

Other14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014
CountrySweden
CityGothenburg
Period26/4/1430/4/14

Fingerprint

Syntactics
Semantics
Experiments

ASJC Scopus subject areas

  • Software

Cite this

Tymoshenko, K., Moschitti, A., & Severyn, A. (2014). Encoding semantic resources in syntactic structures for passage reranking. In 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014 (pp. 664-672). Association for Computational Linguistics (ACL).

Encoding semantic resources in syntactic structures for passage reranking. / Tymoshenko, Kateryna; Moschitti, Alessandro; Severyn, Aliaksei.

14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014. Association for Computational Linguistics (ACL), 2014. p. 664-672.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tymoshenko, K, Moschitti, A & Severyn, A 2014, Encoding semantic resources in syntactic structures for passage reranking. in 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014. Association for Computational Linguistics (ACL), pp. 664-672, 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014, Gothenburg, Sweden, 26/4/14.
Tymoshenko K, Moschitti A, Severyn A. Encoding semantic resources in syntactic structures for passage reranking. In 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014. Association for Computational Linguistics (ACL). 2014. p. 664-672
Tymoshenko, Kateryna ; Moschitti, Alessandro ; Severyn, Aliaksei. / Encoding semantic resources in syntactic structures for passage reranking. 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014. Association for Computational Linguistics (ACL), 2014. pp. 664-672
@inproceedings{7789d5797b46459292d5c6de5ea4c8cf,
title = "Encoding semantic resources in syntactic structures for passage reranking",
abstract = "In this paper, we propose to use semantic knowledge from Wikipedia and large-scale structured knowledge datasets available as Linked Open Data (LOD) for the answer passage reranking task. We represent question and candidate answer passages with pairs of shallow syntactic/semantic trees, whose constituents are connected using LOD. The trees are processed by SVMs and tree kernels, which can automatically exploit tree fragments. The experiments with our SVM rank algorithm on the TREC Question Answering (QA) corpus show that the added relational information highly improves over the state of the art, e.g., about 15.4{\%} of relative improvement in P@1.",
author = "Kateryna Tymoshenko and Alessandro Moschitti and Aliaksei Severyn",
year = "2014",
month = "1",
day = "1",
language = "English",
isbn = "9781632663962",
pages = "664--672",
booktitle = "14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014",
publisher = "Association for Computational Linguistics (ACL)",

}

TY - GEN

T1 - Encoding semantic resources in syntactic structures for passage reranking

AU - Tymoshenko, Kateryna

AU - Moschitti, Alessandro

AU - Severyn, Aliaksei

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this paper, we propose to use semantic knowledge from Wikipedia and large-scale structured knowledge datasets available as Linked Open Data (LOD) for the answer passage reranking task. We represent question and candidate answer passages with pairs of shallow syntactic/semantic trees, whose constituents are connected using LOD. The trees are processed by SVMs and tree kernels, which can automatically exploit tree fragments. The experiments with our SVM rank algorithm on the TREC Question Answering (QA) corpus show that the added relational information highly improves over the state of the art, e.g., about 15.4% of relative improvement in P@1.

AB - In this paper, we propose to use semantic knowledge from Wikipedia and large-scale structured knowledge datasets available as Linked Open Data (LOD) for the answer passage reranking task. We represent question and candidate answer passages with pairs of shallow syntactic/semantic trees, whose constituents are connected using LOD. The trees are processed by SVMs and tree kernels, which can automatically exploit tree fragments. The experiments with our SVM rank algorithm on the TREC Question Answering (QA) corpus show that the added relational information highly improves over the state of the art, e.g., about 15.4% of relative improvement in P@1.

UR - http://www.scopus.com/inward/record.url?scp=84905674833&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905674833&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84905674833

SN - 9781632663962

SP - 664

EP - 672

BT - 14th Conference of the European Chapter of the Association for Computational Linguistics 2014, EACL 2014

PB - Association for Computational Linguistics (ACL)

ER -