Domain-Specific Word Embeddings with Structure Prediction
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Domain-Specific Word Embeddings with Structure Prediction. / Lassner, David; Brandl, Stephanie; Baillot, Anne; Nakajima, Shinichi.
In: Transactions of the Association for Computational Linguistics, Vol. 11, 2023, p. 320-335.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Domain-Specific Word Embeddings with Structure Prediction
AU - Lassner, David
AU - Brandl, Stephanie
AU - Baillot, Anne
AU - Nakajima, Shinichi
N1 - Publisher Copyright: © 2023 Association for Computational Linguistics. Distributed under a CC-BY 4.0 license.
PY - 2023
Y1 - 2023
N2 - Complementary to finding good general word embeddings, an important question for representation learning is to find dynamic word embeddings, for example, across time or domain. Current methods do not offer a way to use or predict information on structure between sub-corpora, time or domain and dynamic embeddings can only be compared after postalignment. We propose novel word embedding methods that provide general word representations for the whole corpus, domainspecific representations for each sub-corpus, sub-corpus structure, and embedding alignment simultaneously. We present an empirical evaluation on New York Times articles and two English Wikipedia datasets with articles on science and philosophy. Our method, called Word2Vec with Structure Prediction (W2VPred), provides better performance than baselines in terms of the general analogy tests, domain-specific analogy tests, and multiple specific word embedding evaluations as well as structure prediction performance when no structure is given a priori. As a use case in the field of Digital Humanities we demonstrate how to raise novel research questions for high literature from the German Text Archive.
AB - Complementary to finding good general word embeddings, an important question for representation learning is to find dynamic word embeddings, for example, across time or domain. Current methods do not offer a way to use or predict information on structure between sub-corpora, time or domain and dynamic embeddings can only be compared after postalignment. We propose novel word embedding methods that provide general word representations for the whole corpus, domainspecific representations for each sub-corpus, sub-corpus structure, and embedding alignment simultaneously. We present an empirical evaluation on New York Times articles and two English Wikipedia datasets with articles on science and philosophy. Our method, called Word2Vec with Structure Prediction (W2VPred), provides better performance than baselines in terms of the general analogy tests, domain-specific analogy tests, and multiple specific word embedding evaluations as well as structure prediction performance when no structure is given a priori. As a use case in the field of Digital Humanities we demonstrate how to raise novel research questions for high literature from the German Text Archive.
U2 - 10.1162/tacl_a_00538
DO - 10.1162/tacl_a_00538
M3 - Journal article
AN - SCOPUS:85153523524
VL - 11
SP - 320
EP - 335
JO - Transactions of the Association for Computational Linguistics
JF - Transactions of the Association for Computational Linguistics
SN - 2307-387X
ER -
ID: 371562185