Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment

Tu Vu; Vered Shwartz

Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment

Tu Vu, Vered Shwartz

CS@BIU

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Scopus citations

Abstract

Supervised distributional methods are applied successfully in lexical entailment, but recent work questioned whether these methods actually learn a relation between two words. Specifically, Levy et al. (2015) claimed that linear classifiers learn only separate properties of each word. We suggest a cheap and easy way to boost the performance of these methods by integrating multiplicative features into commonly used representations. We provide an extensive evaluation with different classifiers and evaluation setups, and suggest a suitable evaluation setup for the task, eliminating biases existing in previous ones.

Original language	English
Title of host publication	NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference
Editors	Malvina Nissim, Jonathan Berant, Alessandro Lenci
Publisher	Association for Computational Linguistics (ACL)
Pages	160-166
Number of pages	7
ISBN (Electronic)	9781948087223
State	Published - 2018
Event	7th Joint Conference on Lexical and Computational Semantics, SEM 2018, co-located with NAACL HLT 2018 - New Orleans, United States Duration: 5 Jun 2018 → 6 Jun 2018

Publication series

Name	NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference

Conference

Conference	7th Joint Conference on Lexical and Computational Semantics, SEM 2018, co-located with NAACL HLT 2018
Country/Territory	United States
City	New Orleans
Period	5/06/18 → 6/06/18

Bibliographical note

Funding Information:
Vered is supported in part by an Intel ICRI-CI grant, the Israel Science Foundation grant 1951/17, the German Research Foundation through the German-Israeli Project Cooperation (DIP, grant DA 1600/1-1), the Clore Scholars Pro-gramme (2017), and the AI2 Key Scientific Challenges Program (2017).

Funding Information:
Vered is supported in part by an Intel ICRI-CI grant, the Israel Science Foundation grant 1951/17, the German Research Foundation through the German-Israeli Project Cooperation (DIP, grant DA 1600/1-1), the Clore Scholars Programme (2017), and the AI2 Key Scientific Challenges Program (2017).

Publisher Copyright:
© 2018 Association for Computational Linguistics.

Cite this

Vu, T., & Shwartz, V. (2018). Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment. In M. Nissim, J. Berant, & A. Lenci (Eds.), NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference (pp. 160-166). (NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference). Association for Computational Linguistics (ACL).

Vu, Tu ; Shwartz, Vered. / Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment. NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference. editor / Malvina Nissim ; Jonathan Berant ; Alessandro Lenci. Association for Computational Linguistics (ACL), 2018. pp. 160-166 (NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference).

@inproceedings{a8ce5cd516884182b1be9f73bf61d3ec,

title = "Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment",

abstract = "Supervised distributional methods are applied successfully in lexical entailment, but recent work questioned whether these methods actually learn a relation between two words. Specifically, Levy et al. (2015) claimed that linear classifiers learn only separate properties of each word. We suggest a cheap and easy way to boost the performance of these methods by integrating multiplicative features into commonly used representations. We provide an extensive evaluation with different classifiers and evaluation setups, and suggest a suitable evaluation setup for the task, eliminating biases existing in previous ones.",

author = "Tu Vu and Vered Shwartz",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics.; 7th Joint Conference on Lexical and Computational Semantics, SEM 2018, co-located with NAACL HLT 2018 ; Conference date: 05-06-2018 Through 06-06-2018",

year = "2018",

language = "אנגלית",

series = "NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference",

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

pages = "160--166",

editor = "Malvina Nissim and Jonathan Berant and Alessandro Lenci",

booktitle = "NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference",

address = "ארצות הברית",

}

Vu, T & Shwartz, V 2018, Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment. in M Nissim, J Berant & A Lenci (eds), NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference. NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference, Association for Computational Linguistics (ACL), pp. 160-166, 7th Joint Conference on Lexical and Computational Semantics, SEM 2018, co-located with NAACL HLT 2018, New Orleans, United States, 5/06/18.

Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment. / Vu, Tu; Shwartz, Vered.
NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference. ed. / Malvina Nissim; Jonathan Berant; Alessandro Lenci. Association for Computational Linguistics (ACL), 2018. p. 160-166 (NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment

AU - Vu, Tu

AU - Shwartz, Vered

PY - 2018

Y1 - 2018

N2 - Supervised distributional methods are applied successfully in lexical entailment, but recent work questioned whether these methods actually learn a relation between two words. Specifically, Levy et al. (2015) claimed that linear classifiers learn only separate properties of each word. We suggest a cheap and easy way to boost the performance of these methods by integrating multiplicative features into commonly used representations. We provide an extensive evaluation with different classifiers and evaluation setups, and suggest a suitable evaluation setup for the task, eliminating biases existing in previous ones.

AB - Supervised distributional methods are applied successfully in lexical entailment, but recent work questioned whether these methods actually learn a relation between two words. Specifically, Levy et al. (2015) claimed that linear classifiers learn only separate properties of each word. We suggest a cheap and easy way to boost the performance of these methods by integrating multiplicative features into commonly used representations. We provide an extensive evaluation with different classifiers and evaluation setups, and suggest a suitable evaluation setup for the task, eliminating biases existing in previous ones.

UR - http://www.scopus.com/inward/record.url?scp=85084045159&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85084045159

T3 - NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference

SP - 160

EP - 166

BT - NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference

A2 - Nissim, Malvina

A2 - Berant, Jonathan

A2 - Lenci, Alessandro

PB - Association for Computational Linguistics (ACL)

T2 - 7th Joint Conference on Lexical and Computational Semantics, SEM 2018, co-located with NAACL HLT 2018

Y2 - 5 June 2018 through 6 June 2018

ER -

Vu T, Shwartz V. Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment. In Nissim M, Berant J, Lenci A, editors, NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference. Association for Computational Linguistics (ACL). 2018. p. 160-166. (NAACL HLT 2018 - Lexical and Computational Semantics, SEM 2018, Proceedings of the 7th Conference).

Integrating Multiplicative Features into Supervised Distributional Methods for Lexical Entailment

Abstract

Publication series

Conference

Bibliographical note

Other files and links

Fingerprint

Cite this