The interplay between lexicography and modern natural language processing: a tale of two marrying worlds

Espinosa-Anke, Luis ORCID: https://orcid.org/0000-0001-6830-9176 2026. The interplay between lexicography and modern natural language processing: a tale of two marrying worlds. Reference Module in Social Sciences, Elsevier, (10.1016/b978-0-323-95504-1.00695-5)

Full text not available from this repository.

Abstract

Lexicography and textual representations (typically in the form of dense vectors also known as embeddings) have had a strong relationship in computational linguistics and natural language processing. There is a significant body of work that has shown how to leverage lexical resources like WordNet for developing word sense disambiguation or ontology learning systems. In fact, in modern works, lexical resources have not been used to directly solve a specific automatic text processing task, but rather, as a source for language models to “pre-train” on, the result of which is often more accurate language models, better at understanding the meaning of linguistic units at various levels, from words to sentences, and in particular, dictionary-relevant items such as definitions, examples of usage or semantic relationships.

Item Type:	Book Section
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Computer Science & Informatics
Publisher:	Elsevier
ISBN:	9780443157851
Last Modified:	19 Jan 2026 13:16
URI:	https://orca.cardiff.ac.uk/id/eprint/184014

Actions (repository staff only)

Edit Item