A study on similarity and relatedness using distributional and WordNet-based approaches
E. Agirre, E. Alfonseca, K. Hall, J. Kravalova, M. Pasca, and A. Soroa. Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, page 19--27. Stroudsburg, PA, USA, Association for Computational Linguistics, (2009)
Abstract
This paper presents and compares WordNet-based and distributional similarity approaches. The strengths and weaknesses of each approach regarding similarity and relatedness tasks are discussed, and a combination is presented. Each of our methods independently provide the best results in their class on the RG and WordSim353 datasets, and a supervised combination of them yields the best published results on all datasets. Finally, we pioneer cross-lingual similarity, showing that our methods are easily adapted for a cross-lingual task with minor losses.
Description
A study on similarity and relatedness using distributional and WordNet-based approaches
%0 Conference Paper
%1 agirre2009study
%A Agirre, Eneko
%A Alfonseca, Enrique
%A Hall, Keith
%A Kravalova, Jana
%A Pasca, Marius
%A Soroa, Aitor
%B Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics
%C Stroudsburg, PA, USA
%D 2009
%I Association for Computational Linguistics
%K matching relatedness semantic semantic-measure-similarity sts
%P 19--27
%T A study on similarity and relatedness using distributional and WordNet-based approaches
%U https://www.aclweb.org/anthology/N09-1003
%X This paper presents and compares WordNet-based and distributional similarity approaches. The strengths and weaknesses of each approach regarding similarity and relatedness tasks are discussed, and a combination is presented. Each of our methods independently provide the best results in their class on the RG and WordSim353 datasets, and a supervised combination of them yields the best published results on all datasets. Finally, we pioneer cross-lingual similarity, showing that our methods are easily adapted for a cross-lingual task with minor losses.
%@ 978-1-932432-41-1
@inproceedings{agirre2009study,
abstract = {This paper presents and compares WordNet-based and distributional similarity approaches. The strengths and weaknesses of each approach regarding similarity and relatedness tasks are discussed, and a combination is presented. Each of our methods independently provide the best results in their class on the RG and WordSim353 datasets, and a supervised combination of them yields the best published results on all datasets. Finally, we pioneer cross-lingual similarity, showing that our methods are easily adapted for a cross-lingual task with minor losses.},
acmid = {1620758},
added-at = {2019-09-30T21:22:24.000+0200},
address = {Stroudsburg, PA, USA},
author = {Agirre, Eneko and Alfonseca, Enrique and Hall, Keith and Kravalova, Jana and Pa\c{s}ca, Marius and Soroa, Aitor},
biburl = {https://www.bibsonomy.org/bibtex/2e91cce045a1d323c0012027f04771733/blasp},
booktitle = {Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics},
description = {A study on similarity and relatedness using distributional and WordNet-based approaches},
interhash = {35326b1cfd5cde92744c22b981c84b23},
intrahash = {e91cce045a1d323c0012027f04771733},
isbn = {978-1-932432-41-1},
keywords = {matching relatedness semantic semantic-measure-similarity sts},
location = {Boulder, Colorado},
numpages = {9},
pages = {19--27},
publisher = {Association for Computational Linguistics},
series = {NAACL '09},
timestamp = {2019-09-30T21:22:24.000+0200},
title = {A study on similarity and relatedness using distributional and WordNet-based approaches},
url = {https://www.aclweb.org/anthology/N09-1003},
year = 2009
}