Visualization Methods for Diachronic Semantic Shift

Raef Kazi, Alessandra Amato, Shenghui Wang, Doina Bucur

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

26 Downloads (Pure)

Abstract

The meaning and usage of a concept or a word changes over time. These diachronic semantic shifts reflect the change of societal and cultural consensus as well as the evolution of science. The availability of large-scale corpora and recent success in language models have enabled researchers to analyze semantic shifts in great detail. However, current research lacks intuitive ways of presenting diachronic semantic shifts and making them comprehensive. In this paper, we study the PubMed dataset and compute semantic shifts across six decades. We develop three visualization methods that can show, given a root word: the temporal change in its linguistic context, word re-occurrence, degree of similarity, time continuity, and separate trends per geographic location. We also propose a taxonomy that classifies visualization methods for diachronic semantic shifts with respect to different purposes.

Original languageEnglish
Title of host publicationProceedings of the Third Workshop on Scholarly Document Processing
PublisherAssociation for Computational Linguistics (ACL)
Pages89-94
Number of pages6
Publication statusPublished - Oct 2022
Event3rd Workshop on Scholarly Document Processing, SDP 2022 - Gyeongju, Korea, Republic of
Duration: 17 Oct 202217 Oct 2022
Conference number: 3
https://sdproc.org/2022/

Publication series

NameProceedings - International Conference on Computational Linguistics, COLING
ISSN (Print)2951-2093

Workshop

Workshop3rd Workshop on Scholarly Document Processing, SDP 2022
Abbreviated titleSDP 2022
Country/TerritoryKorea, Republic of
CityGyeongju
Period17/10/2217/10/22
Otherto be held at COLING 2022 (October 12-17, 2022)
Internet address

Fingerprint

Dive into the research topics of 'Visualization Methods for Diachronic Semantic Shift'. Together they form a unique fingerprint.

Cite this