A rank-based similarity metric for word embeddings

Enrico Santus; Hongmin Wang; Emmanuele Chersoni; Yue Zhang

doi:10.18653/v1/p18-2088

A rank-based similarity metric for word embeddings

Enrico Santus
, Hongmin Wang
, Emmanuele Chersoni
, Yue Zhang

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

11 Citations (Scopus)

Abstract

Word Embeddings (WE) have recently imposed themselves as a standard for representing word meaning in NLP. Semantic similarity between word pairs has become the most common evaluation benchmark for these representations, with vector cosine being typically used as the only similarity metric. In this paper, we report experiments with a rank-based metric for WE, which performs comparably to vector cosine in similarity estimation and outperforms it in the recently-introduced and challenging task of outlier detection, thus suggesting that rank-based measures can improve clustering quality.¹

Original language	English
Title of host publication	Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Editors	Iryna Gurevych, Yusuke Miyao
Publisher	Association for Computational Linguistics (ACL)
Pages	552-557
Number of pages	6
ISBN (Electronic)	9781948087346
DOIs	https://doi.org/10.18653/v1/p18-2088
Publication status	Published - Jul 2018
Externally published	Yes
Event	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 - Melbourne, Australia Duration: 15 Jul 2018 → 20 Jul 2018

Publication series

Name	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
Volume	2

Conference

Conference	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018
Country/Territory	Australia
City	Melbourne
Period	15/07/18 → 20/07/18

ASJC Scopus subject areas

Software
Computational Theory and Mathematics

More information

10.18653/v1/p18-2088

Cite this

Santus, E., Wang, H., Chersoni, E., & Zhang, Y. (2018). A rank-based similarity metric for word embeddings. In I. Gurevych, & Y. Miyao (Eds.), Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (pp. 552-557). (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/p18-2088

Santus, Enrico ; Wang, Hongmin ; Chersoni, Emmanuele et al. / A rank-based similarity metric for word embeddings. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). editor / Iryna Gurevych ; Yusuke Miyao. Association for Computational Linguistics (ACL), 2018. pp. 552-557 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)).

@inproceedings{b5288270a7e94ddc88c97c72c8a55324,

title = "A rank-based similarity metric for word embeddings",

abstract = "Word Embeddings (WE) have recently imposed themselves as a standard for representing word meaning in NLP. Semantic similarity between word pairs has become the most common evaluation benchmark for these representations, with vector cosine being typically used as the only similarity metric. In this paper, we report experiments with a rank-based metric for WE, which performs comparably to vector cosine in similarity estimation and outperforms it in the recently-introduced and challenging task of outlier detection, thus suggesting that rank-based measures can improve clustering quality.1",

author = "Enrico Santus and Hongmin Wang and Emmanuele Chersoni and Yue Zhang",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics; 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 ; Conference date: 15-07-2018 Through 20-07-2018",

year = "2018",

month = jul,

doi = "10.18653/v1/p18-2088",

language = "English",

series = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)",

publisher = "Association for Computational Linguistics (ACL)",

pages = "552--557",

editor = "Iryna Gurevych and Yusuke Miyao",

booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",

}

Santus, E, Wang, H, Chersoni, E & Zhang, Y 2018, A rank-based similarity metric for word embeddings. in I Gurevych & Y Miyao (eds), Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), vol. 2, Association for Computational Linguistics (ACL), pp. 552-557, 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018, Melbourne, Australia, 15/07/18. https://doi.org/10.18653/v1/p18-2088

A rank-based similarity metric for word embeddings. / Santus, Enrico; Wang, Hongmin; Chersoni, Emmanuele et al.
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). ed. / Iryna Gurevych; Yusuke Miyao. Association for Computational Linguistics (ACL), 2018. p. 552-557 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - A rank-based similarity metric for word embeddings

AU - Santus, Enrico

AU - Wang, Hongmin

AU - Chersoni, Emmanuele

AU - Zhang, Yue

PY - 2018/7

Y1 - 2018/7

N2 - Word Embeddings (WE) have recently imposed themselves as a standard for representing word meaning in NLP. Semantic similarity between word pairs has become the most common evaluation benchmark for these representations, with vector cosine being typically used as the only similarity metric. In this paper, we report experiments with a rank-based metric for WE, which performs comparably to vector cosine in similarity estimation and outperforms it in the recently-introduced and challenging task of outlier detection, thus suggesting that rank-based measures can improve clustering quality.1

AB - Word Embeddings (WE) have recently imposed themselves as a standard for representing word meaning in NLP. Semantic similarity between word pairs has become the most common evaluation benchmark for these representations, with vector cosine being typically used as the only similarity metric. In this paper, we report experiments with a rank-based metric for WE, which performs comparably to vector cosine in similarity estimation and outperforms it in the recently-introduced and challenging task of outlier detection, thus suggesting that rank-based measures can improve clustering quality.1

UR - https://www.scopus.com/pages/publications/85063129461

U2 - 10.18653/v1/p18-2088

DO - 10.18653/v1/p18-2088

M3 - Conference article published in proceeding or book

AN - SCOPUS:85063129461

T3 - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)

SP - 552

EP - 557

BT - Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

A2 - Gurevych, Iryna

A2 - Miyao, Yusuke

PB - Association for Computational Linguistics (ACL)

T2 - 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018

Y2 - 15 July 2018 through 20 July 2018

ER -

Santus E, Wang H, Chersoni E, Zhang Y. A rank-based similarity metric for word embeddings. In Gurevych I, Miyao Y, editors, Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). Association for Computational Linguistics (ACL). 2018. p. 552-557. (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)). doi: 10.18653/v1/p18-2088

A rank-based similarity metric for word embeddings

Abstract

Publication series

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this