A distributional study of negated adjectives and antonyms

Open Access
Authors
Publication date 2018
Host editors
  • E. Cabrio
  • A. Mazzei
  • F. Tamburini
Book title Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018)
Book subtitle Torino, Italy, December 10-12, 2018
Series CEUR Workshop Proceedings
Event 5th Italian Conference on Computational Linguistics
Number of pages 7
Publisher Aachen: CEUR-WS
Organisations
  • Interfacultary Research - Institute for Logic, Language and Computation (ILLC)
Abstract In this paper, we investigate the relation between negated adjectives and antonyms in English using Distributional Semantics methods. Results show that, on the basis of contexts of use, a negated adjective (e.g., not cold) is typically more similar to the adjective itself (cold) than to its antonym (hot); such effect is less strong for antonyms derived by affixation (e.g., happy - unhappy).
Document type Conference contribution
Language English
Published at http://ceur-ws.org/Vol-2253/paper03.pdf
Other links http://ceur-ws.org/Vol-2253/
Downloads
paper03 (Final published version)
Permalink to this page
Back