GASC: Genre-Aware Semantic Change for Ancient Greek

Valerio Perrone, Marco Palma, Simon Hengchen, Alessandro Vatri, Jim Q. Smith, Barbara McGillivray

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

Abstract

Word meaning changes over time, depending on linguistic and extra-linguistic factors. Associating a word's correct meaning in its historical context is a central challenge in diachronic research, and is relevant to a range of NLP tasks, including information retrieval and semantic search in historical texts. Bayesian models for semantic change have emerged as a powerful tool to address this challenge, providing explicit and interpretable representations of semantic change phenomena. However, while corpora typically come with rich metadata, existing models are limited by their inability to exploit contextual information (such as text genre) beyond the document time-stamp. This is particularly critical in the case of ancient languages, where lack of data and long diachronic span make it harder to draw a clear distinction between polysemy (the fact that a word has several senses) and semantic change (the process of acquiring, losing, or changing senses), and current systems perform poorly on these languages. We develop GASC, a dynamic semantic change model that leverages categorical metadata about the texts' genre to boost inference and uncover the evolution of meanings in Ancient Greek corpora. In a new evaluation framework, our model achieves improved predictive performance compared to the state of the art.
Original languageEnglish
Title of host publicationThe 1st InternationalWorkshop on Computational Approaches to Historical Language Change : Proceedings of the Workshop
Number of pages11
Place of PublicationStroudsburg
PublisherACL
Publication dateJul 2019
Pages56-66
ISBN (Electronic)978-1-950737-31-4
Publication statusPublished - Jul 2019
MoE publication typeA4 Article in conference proceedings
EventInternational Workshop on Computational
Approaches to Historical Language Change
- Florence, Italy
Duration: 2 Aug 20192 Aug 2019
Conference number: 1

Fields of Science

  • 112 Statistics and probability
  • 6121 Languages

Cite this

Perrone, V., Palma, M., Hengchen, S., Vatri, A., Smith, J. Q., & McGillivray, B. (2019). GASC: Genre-Aware Semantic Change for Ancient Greek. In The 1st InternationalWorkshop on Computational Approaches to Historical Language Change: Proceedings of the Workshop (pp. 56-66). Stroudsburg: ACL.
Perrone, Valerio ; Palma, Marco ; Hengchen, Simon ; Vatri, Alessandro ; Smith, Jim Q. ; McGillivray, Barbara. / GASC: Genre-Aware Semantic Change for Ancient Greek. The 1st InternationalWorkshop on Computational Approaches to Historical Language Change: Proceedings of the Workshop. Stroudsburg : ACL, 2019. pp. 56-66
@inproceedings{128e4b72c0284a51a59ba6a6db3ae9a5,
title = "GASC: Genre-Aware Semantic Change for Ancient Greek",
abstract = "Word meaning changes over time, depending on linguistic and extra-linguistic factors. Associating a word's correct meaning in its historical context is a central challenge in diachronic research, and is relevant to a range of NLP tasks, including information retrieval and semantic search in historical texts. Bayesian models for semantic change have emerged as a powerful tool to address this challenge, providing explicit and interpretable representations of semantic change phenomena. However, while corpora typically come with rich metadata, existing models are limited by their inability to exploit contextual information (such as text genre) beyond the document time-stamp. This is particularly critical in the case of ancient languages, where lack of data and long diachronic span make it harder to draw a clear distinction between polysemy (the fact that a word has several senses) and semantic change (the process of acquiring, losing, or changing senses), and current systems perform poorly on these languages. We develop GASC, a dynamic semantic change model that leverages categorical metadata about the texts' genre to boost inference and uncover the evolution of meanings in Ancient Greek corpora. In a new evaluation framework, our model achieves improved predictive performance compared to the state of the art.",
keywords = "112 Statistics and probability, 6121 Languages",
author = "Valerio Perrone and Marco Palma and Simon Hengchen and Alessandro Vatri and Smith, {Jim Q.} and Barbara McGillivray",
year = "2019",
month = "7",
language = "English",
pages = "56--66",
booktitle = "The 1st InternationalWorkshop on Computational Approaches to Historical Language Change",
publisher = "ACL",
address = "United States",

}

Perrone, V, Palma, M, Hengchen, S, Vatri, A, Smith, JQ & McGillivray, B 2019, GASC: Genre-Aware Semantic Change for Ancient Greek. in The 1st InternationalWorkshop on Computational Approaches to Historical Language Change: Proceedings of the Workshop. ACL, Stroudsburg, pp. 56-66, International Workshop on Computational
Approaches to Historical Language Change, Florence, Italy, 02/08/2019.

GASC: Genre-Aware Semantic Change for Ancient Greek. / Perrone, Valerio; Palma, Marco; Hengchen, Simon; Vatri, Alessandro; Smith, Jim Q.; McGillivray, Barbara.

The 1st InternationalWorkshop on Computational Approaches to Historical Language Change: Proceedings of the Workshop. Stroudsburg : ACL, 2019. p. 56-66.

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

TY - GEN

T1 - GASC: Genre-Aware Semantic Change for Ancient Greek

AU - Perrone, Valerio

AU - Palma, Marco

AU - Hengchen, Simon

AU - Vatri, Alessandro

AU - Smith, Jim Q.

AU - McGillivray, Barbara

PY - 2019/7

Y1 - 2019/7

N2 - Word meaning changes over time, depending on linguistic and extra-linguistic factors. Associating a word's correct meaning in its historical context is a central challenge in diachronic research, and is relevant to a range of NLP tasks, including information retrieval and semantic search in historical texts. Bayesian models for semantic change have emerged as a powerful tool to address this challenge, providing explicit and interpretable representations of semantic change phenomena. However, while corpora typically come with rich metadata, existing models are limited by their inability to exploit contextual information (such as text genre) beyond the document time-stamp. This is particularly critical in the case of ancient languages, where lack of data and long diachronic span make it harder to draw a clear distinction between polysemy (the fact that a word has several senses) and semantic change (the process of acquiring, losing, or changing senses), and current systems perform poorly on these languages. We develop GASC, a dynamic semantic change model that leverages categorical metadata about the texts' genre to boost inference and uncover the evolution of meanings in Ancient Greek corpora. In a new evaluation framework, our model achieves improved predictive performance compared to the state of the art.

AB - Word meaning changes over time, depending on linguistic and extra-linguistic factors. Associating a word's correct meaning in its historical context is a central challenge in diachronic research, and is relevant to a range of NLP tasks, including information retrieval and semantic search in historical texts. Bayesian models for semantic change have emerged as a powerful tool to address this challenge, providing explicit and interpretable representations of semantic change phenomena. However, while corpora typically come with rich metadata, existing models are limited by their inability to exploit contextual information (such as text genre) beyond the document time-stamp. This is particularly critical in the case of ancient languages, where lack of data and long diachronic span make it harder to draw a clear distinction between polysemy (the fact that a word has several senses) and semantic change (the process of acquiring, losing, or changing senses), and current systems perform poorly on these languages. We develop GASC, a dynamic semantic change model that leverages categorical metadata about the texts' genre to boost inference and uncover the evolution of meanings in Ancient Greek corpora. In a new evaluation framework, our model achieves improved predictive performance compared to the state of the art.

KW - 112 Statistics and probability

KW - 6121 Languages

M3 - Conference contribution

SP - 56

EP - 66

BT - The 1st InternationalWorkshop on Computational Approaches to Historical Language Change

PB - ACL

CY - Stroudsburg

ER -

Perrone V, Palma M, Hengchen S, Vatri A, Smith JQ, McGillivray B. GASC: Genre-Aware Semantic Change for Ancient Greek. In The 1st InternationalWorkshop on Computational Approaches to Historical Language Change: Proceedings of the Workshop. Stroudsburg: ACL. 2019. p. 56-66