Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change

Haim Dubossarsky, Simon Hengchen, Nina Tahmasebi, Dominik Schlechtweg

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

State-of-the-art models of lexical semantic change detection suffer from noise stemming from vector space alignment. We have empirically tested the Temporal Referencing method for lexical semantic change and show that, by avoiding alignment, it is less affected by this noise. We show that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset. We introduce a principled way to simulate lexical semantic change and systematically control for possible biases.
Original languageEnglish
Title of host publicationThe 57th Annual Meeting of the Association for Computational Linguistics (ACL2019) : Proceedings of the Conference
EditorsAnna Korhonen, David Traum, Lluís Màrquez
Number of pages14
Place of PublicationStroudsburg
PublisherACL
Publication dateJul 2019
Pages457-470
ISBN (Electronic)978-1-950737-48-2
Publication statusPublished - Jul 2019
MoE publication typeA4 Article in conference proceedings
EventAnnual Meeting of the Association for Computational Linguistics - Florence, Italy
Duration: 28 Jul 20192 Aug 2019
Conference number: 57
http://www.acl2019.org/EN/index.xhtml

Bibliographical note

Code produced for this paper is available at: https://github.com/Garrafao/TemporalReferencing

Fields of Science

  • 113 Computer and information sciences
  • Natural language processing
  • Computational linguistics
  • 6121 Languages

Cite this

Dubossarsky, H., Hengchen, S., Tahmasebi, N., & Schlechtweg, D. (2019). Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change. In A. Korhonen, D. Traum, & L. Màrquez (Eds.), The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019): Proceedings of the Conference (pp. 457-470). Stroudsburg: ACL.
Dubossarsky, Haim ; Hengchen, Simon ; Tahmasebi, Nina ; Schlechtweg, Dominik. / Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change. The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019): Proceedings of the Conference. editor / Anna Korhonen ; David Traum ; Lluís Màrquez . Stroudsburg : ACL, 2019. pp. 457-470
@inproceedings{50ad383b55c54e6e90fb1841f6413480,
title = "Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change",
abstract = "State-of-the-art models of lexical semantic change detection suffer from noise stemming from vector space alignment. We have empirically tested the Temporal Referencing method for lexical semantic change and show that, by avoiding alignment, it is less affected by this noise. We show that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset. We introduce a principled way to simulate lexical semantic change and systematically control for possible biases.",
keywords = "113 Computer and information sciences, Natural language processing, Computational linguistics, 6121 Languages",
author = "Haim Dubossarsky and Simon Hengchen and Nina Tahmasebi and Dominik Schlechtweg",
note = "Code produced for this paper is available at: https://github.com/Garrafao/TemporalReferencing",
year = "2019",
month = "7",
language = "English",
pages = "457--470",
editor = "Anna Korhonen and Traum, {David } and {M{\`a}rquez }, {Llu{\'i}s }",
booktitle = "The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019)",
publisher = "ACL",
address = "United States",

}

Dubossarsky, H, Hengchen, S, Tahmasebi, N & Schlechtweg, D 2019, Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change. in A Korhonen, D Traum & L Màrquez (eds), The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019): Proceedings of the Conference. ACL, Stroudsburg, pp. 457-470, Annual Meeting of the Association for Computational Linguistics, Florence, Italy, 28/07/2019.

Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change. / Dubossarsky, Haim; Hengchen, Simon; Tahmasebi, Nina; Schlechtweg, Dominik.

The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019): Proceedings of the Conference. ed. / Anna Korhonen; David Traum; Lluís Màrquez . Stroudsburg : ACL, 2019. p. 457-470.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change

AU - Dubossarsky, Haim

AU - Hengchen, Simon

AU - Tahmasebi, Nina

AU - Schlechtweg, Dominik

N1 - Code produced for this paper is available at: https://github.com/Garrafao/TemporalReferencing

PY - 2019/7

Y1 - 2019/7

N2 - State-of-the-art models of lexical semantic change detection suffer from noise stemming from vector space alignment. We have empirically tested the Temporal Referencing method for lexical semantic change and show that, by avoiding alignment, it is less affected by this noise. We show that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset. We introduce a principled way to simulate lexical semantic change and systematically control for possible biases.

AB - State-of-the-art models of lexical semantic change detection suffer from noise stemming from vector space alignment. We have empirically tested the Temporal Referencing method for lexical semantic change and show that, by avoiding alignment, it is less affected by this noise. We show that, trained on a diachronic corpus, the skip-gram with negative sampling architecture with temporal referencing outperforms alignment models on a synthetic task as well as a manual testset. We introduce a principled way to simulate lexical semantic change and systematically control for possible biases.

KW - 113 Computer and information sciences

KW - Natural language processing

KW - Computational linguistics

KW - 6121 Languages

M3 - Conference contribution

SP - 457

EP - 470

BT - The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019)

A2 - Korhonen, Anna

A2 - Traum, David

A2 - Màrquez , Lluís

PB - ACL

CY - Stroudsburg

ER -

Dubossarsky H, Hengchen S, Tahmasebi N, Schlechtweg D. Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Change. In Korhonen A, Traum D, Màrquez L, editors, The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019): Proceedings of the Conference. Stroudsburg: ACL. 2019. p. 457-470