Cultural Shift or Linguistic Drift? Comparing Two Computational Measures of Semantic Change
W. Hamilton, J. Leskovec, and D. Jurafsky. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, page 2116--2121. Austin, Texas, Association for Computational Linguistics, (2016)
DOI: 10.18653/v1/D16-1229
Abstract
Words shift in meaning for many reasons, including cultural factors like new technologies and regular linguistic processes like subjectification. Understanding the evolution of language and culture requires disentangling these underlying causes. Here we show how two different distributional measures can be used to detect two different types of semantic change. The first measure, which has been used in many previous works, analyzes global shifts in a word’s distributional semantics; it is sensitive to changes due to regular processes of linguistic drift, such as the semantic generalization of promise (“I promise.”→“It promised to be exciting.”). The second measure, which we develop here, focuses on local changes to a word’s nearest semantic neighbors; it is more sensitive to cultural shifts, such as the change in the meaning of cell (“prison cell” → “cell phone”). Comparing measurements made by these two methods allows researchers to determine whether changes are more cultural or linguistic in nature, a distinction that is essential for work in the digital humanities and historical linguistics.
Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing
year
2016
pages
2116--2121
publisher
Association for Computational Linguistics
shorttitle
Cultural Shift or Linguistic Drift?
language
en
file
Hamilton et al - Cultural Shift or Linguistic Drift ~ Comparing Two Computational Measures of Semantic Change.pdf:C\:\\Users\\Admin\\Documents\\Research\\_Paperbase\\Word Embeddings\\Hamilton et al - Cultural Shift or Linguistic Drift ~ Comparing Two Computational Measures of Semantic Change.pdf:application/pdf
%0 Conference Paper
%1 hamilton_cultural_2016
%A Hamilton, William L.
%A Leskovec, Jure
%A Jurafsky, Dan
%B Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing
%C Austin, Texas
%D 2016
%I Association for Computational Linguistics
%K Embedding_Variability Word_Embeddings
%P 2116--2121
%R 10.18653/v1/D16-1229
%T Cultural Shift or Linguistic Drift? Comparing Two Computational Measures of Semantic Change
%U http://aclweb.org/anthology/D16-1229
%X Words shift in meaning for many reasons, including cultural factors like new technologies and regular linguistic processes like subjectification. Understanding the evolution of language and culture requires disentangling these underlying causes. Here we show how two different distributional measures can be used to detect two different types of semantic change. The first measure, which has been used in many previous works, analyzes global shifts in a word’s distributional semantics; it is sensitive to changes due to regular processes of linguistic drift, such as the semantic generalization of promise (“I promise.”→“It promised to be exciting.”). The second measure, which we develop here, focuses on local changes to a word’s nearest semantic neighbors; it is more sensitive to cultural shifts, such as the change in the meaning of cell (“prison cell” → “cell phone”). Comparing measurements made by these two methods allows researchers to determine whether changes are more cultural or linguistic in nature, a distinction that is essential for work in the digital humanities and historical linguistics.
@inproceedings{hamilton_cultural_2016,
abstract = {Words shift in meaning for many reasons, including cultural factors like new technologies and regular linguistic processes like subjectification. Understanding the evolution of language and culture requires disentangling these underlying causes. Here we show how two different distributional measures can be used to detect two different types of semantic change. The first measure, which has been used in many previous works, analyzes global shifts in a word’s distributional semantics; it is sensitive to changes due to regular processes of linguistic drift, such as the semantic generalization of promise (“I promise.”→“It promised to be exciting.”). The second measure, which we develop here, focuses on local changes to a word’s nearest semantic neighbors; it is more sensitive to cultural shifts, such as the change in the meaning of cell (“prison cell” → “cell phone”). Comparing measurements made by these two methods allows researchers to determine whether changes are more cultural or linguistic in nature, a distinction that is essential for work in the digital humanities and historical linguistics.},
added-at = {2020-02-21T16:09:44.000+0100},
address = {Austin, Texas},
author = {Hamilton, William L. and Leskovec, Jure and Jurafsky, Dan},
biburl = {https://www.bibsonomy.org/bibtex/26e79d02469860f82f1f7dd0b7c084809/tschumacher},
booktitle = {Proceedings of the 2016 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},
doi = {10.18653/v1/D16-1229},
file = {Hamilton et al - Cultural Shift or Linguistic Drift ~ Comparing Two Computational Measures of Semantic Change.pdf:C\:\\Users\\Admin\\Documents\\Research\\_Paperbase\\Word Embeddings\\Hamilton et al - Cultural Shift or Linguistic Drift ~ Comparing Two Computational Measures of Semantic Change.pdf:application/pdf},
interhash = {c6c614275afa4ff333a66bf420cde2e9},
intrahash = {6e79d02469860f82f1f7dd0b7c084809},
keywords = {Embedding_Variability Word_Embeddings},
language = {en},
pages = {2116--2121},
publisher = {Association for Computational Linguistics},
shorttitle = {Cultural {Shift} or {Linguistic} {Drift}?},
timestamp = {2020-02-21T16:09:44.000+0100},
title = {Cultural {Shift} or {Linguistic} {Drift}? {Comparing} {Two} {Computational} {Measures} of {Semantic} {Change}},
url = {http://aclweb.org/anthology/D16-1229},
urldate = {2019-12-11},
year = 2016
}