PageRanking WordNet synsets: An application to opinion mining

Andrea Esuli, Fabrizio Sebastiani

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

101 Citations (Scopus)

Abstract

This paper presents an application of PageRank, a random-walk model originally devised for ranking Web search results, to ranking WordNet synsets in terms of how strongly they possess a given semantic property. The semantic properties we use for exemplifying the approach are positivity and negativity, two properties of central importance in sentiment analysis. The idea derives from the observation that WordNet may be seen as a graph in which synsets are connected through the binary relation "a term belonging to synset s k occurs in the gloss of synset s i", and on the hypothesis that this relation may be viewed as a transmitter of such semantic properties. The data for this relation can be obtained from eXtended WordNet, a publicly available sensedisambiguated version of WordNet. We argue that this relation is structurally akin to the relation between hyperlinkedWeb pages, and thus lends itself to PageRank analysis. We report experimental results supporting our intuitions.

Original languageEnglish
Title of host publicationACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics
Pages424-431
Number of pages8
Publication statusPublished - 2007
Externally publishedYes
Event45th Annual Meeting of the Association for Computational Linguistics, ACL 2007 - Prague
Duration: 23 Jun 200730 Jun 2007

Other

Other45th Annual Meeting of the Association for Computational Linguistics, ACL 2007
CityPrague
Period23/6/0730/6/07

Fingerprint

semantics
ranking
gloss
intuition
WordNet
Semantic Properties
Ranking

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Esuli, A., & Sebastiani, F. (2007). PageRanking WordNet synsets: An application to opinion mining. In ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics (pp. 424-431)

PageRanking WordNet synsets : An application to opinion mining. / Esuli, Andrea; Sebastiani, Fabrizio.

ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics. 2007. p. 424-431.

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

Esuli, A & Sebastiani, F 2007, PageRanking WordNet synsets: An application to opinion mining. in ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics. pp. 424-431, 45th Annual Meeting of the Association for Computational Linguistics, ACL 2007, Prague, 23/6/07.
Esuli A, Sebastiani F. PageRanking WordNet synsets: An application to opinion mining. In ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics. 2007. p. 424-431
Esuli, Andrea ; Sebastiani, Fabrizio. / PageRanking WordNet synsets : An application to opinion mining. ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics. 2007. pp. 424-431
@inproceedings{c3f96bde7e4545f4907c6d9005e70837,
title = "PageRanking WordNet synsets: An application to opinion mining",
abstract = "This paper presents an application of PageRank, a random-walk model originally devised for ranking Web search results, to ranking WordNet synsets in terms of how strongly they possess a given semantic property. The semantic properties we use for exemplifying the approach are positivity and negativity, two properties of central importance in sentiment analysis. The idea derives from the observation that WordNet may be seen as a graph in which synsets are connected through the binary relation {"}a term belonging to synset s k occurs in the gloss of synset s i{"}, and on the hypothesis that this relation may be viewed as a transmitter of such semantic properties. The data for this relation can be obtained from eXtended WordNet, a publicly available sensedisambiguated version of WordNet. We argue that this relation is structurally akin to the relation between hyperlinkedWeb pages, and thus lends itself to PageRank analysis. We report experimental results supporting our intuitions.",
author = "Andrea Esuli and Fabrizio Sebastiani",
year = "2007",
language = "English",
isbn = "9781932432862",
pages = "424--431",
booktitle = "ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics",

}

TY - GEN

T1 - PageRanking WordNet synsets

T2 - An application to opinion mining

AU - Esuli, Andrea

AU - Sebastiani, Fabrizio

PY - 2007

Y1 - 2007

N2 - This paper presents an application of PageRank, a random-walk model originally devised for ranking Web search results, to ranking WordNet synsets in terms of how strongly they possess a given semantic property. The semantic properties we use for exemplifying the approach are positivity and negativity, two properties of central importance in sentiment analysis. The idea derives from the observation that WordNet may be seen as a graph in which synsets are connected through the binary relation "a term belonging to synset s k occurs in the gloss of synset s i", and on the hypothesis that this relation may be viewed as a transmitter of such semantic properties. The data for this relation can be obtained from eXtended WordNet, a publicly available sensedisambiguated version of WordNet. We argue that this relation is structurally akin to the relation between hyperlinkedWeb pages, and thus lends itself to PageRank analysis. We report experimental results supporting our intuitions.

AB - This paper presents an application of PageRank, a random-walk model originally devised for ranking Web search results, to ranking WordNet synsets in terms of how strongly they possess a given semantic property. The semantic properties we use for exemplifying the approach are positivity and negativity, two properties of central importance in sentiment analysis. The idea derives from the observation that WordNet may be seen as a graph in which synsets are connected through the binary relation "a term belonging to synset s k occurs in the gloss of synset s i", and on the hypothesis that this relation may be viewed as a transmitter of such semantic properties. The data for this relation can be obtained from eXtended WordNet, a publicly available sensedisambiguated version of WordNet. We argue that this relation is structurally akin to the relation between hyperlinkedWeb pages, and thus lends itself to PageRank analysis. We report experimental results supporting our intuitions.

UR - http://www.scopus.com/inward/record.url?scp=84860502106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860502106&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84860502106

SN - 9781932432862

SP - 424

EP - 431

BT - ACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics

ER -