Elamur Research⌕
☰
home / papers / 1991 / Co-occurrences of Antonymous Adjectives and Their Contexts

Co-occurrences of Antonymous Adjectives and Their Contexts

Paper · 1991 · John S. Justeson and Slava M. Katz

Summary

The paper asks how people come to link antonymous adjectives such as big–little. It tests the claim by Charles and Miller that this link forms because antonyms appear together in the same sentence, rather than because one can replace the other in a given context. Justeson and Katz counted sentence-level co-occurrences in the part-of-speech-tagged Brown Corpus and checked the results against the much larger, untagged APHB Corpus. They examined three groups: the Deese antonym pairs, other antonym pairs containing a high-frequency adjective, and a large set of morphologically derived opposites (e.g., happy–unhappy). Expected counts and significance came from a hypergeometric model. In all three groups, antonyms shared a sentence roughly ten times more often than chance would predict. When antonyms did co-occur, they typically filled the same slot in two otherwise repeated phrases. The authors argue that this pattern merges both hypotheses: learning happens through co-occurrence, and substitution within repeated frames makes that learning especially effective. The account also links antonym learning to how words of the same part of speech become associated.

Key points

  • Data: The main corpus is the Brown Corpus (1,000,000 words, 500 extracts of about 2,000 words each, 15 genres). After headline pseudo-sentences were dropped, 54,717 sentences remained. The 25,000,000-word APHB Corpus (American Publishing House for the Blind) served as a check.
  • Baseline pairs: The baseline comes from Deese's word-association work on the 278 most frequent Thorndike-Lorge adjectives, which yielded 34 reciprocal antonym pairs plus 5 near-reciprocal ones. Of these, 35 pairs could be tested; 4 pairs, such as few–many, were excluded because of Brown tagging.
  • Main result: Observed-to-expected co-occurrence ratios were all above 2:1 and averaged about 10:1. Possible sentence-splitting errors were estimated to shift these ratios by only about 3%.
  • Statistics: Expected counts come from the hypergeometric formula n₁n₂/N. For the 346 morphological antonym pairs, of which 48 co-occurred, the significance of the number of co-occurring pairs was estimated with 1,000,000 simulation trials.
  • Mechanism: Antonyms usually swap places within otherwise word-for-word repeated clauses or phrases. This provides a restricted form of substitutability inside the co-occurrence hypothesis.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🔬IBM Research - Thomas J. Watson Research CenterYorktown Heights · United States53 works · 49,193 citations

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

Cite

Cite as John S. Justeson and Slava M. Katz. 1991. Co-occurrences of Antonymous Adjectives and Their Contexts. Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work.

@article{justeson1991co,
  title = {{Co-occurrences of Antonymous Adjectives and Their Contexts}},
  author = {John S. Justeson and Slava M. Katz},
  journal = {Computational Linguistics},
  year = {1991},
  doi = {10.5555/971738.971739},
  url = {https://doi.org/10.5555/971738.971739}
}

References

14 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Charles, Walter G. and Miller, George A. (1989). "Contexts of antonymous adjectives." Applied Psycholinguistics, 10(3):357-75. find
  2. Church, Kenneth W. and Hanks, Patrick (1989). "Word association norms, mutual information, and lexicography." Proceedings, 27th Annual Meeting of the (:)
  3. Association for Computational Linguistics, Vancouver, 76-83. find
  4. Deese, James E. (1962). "Form class and the determinants of association." Journal of Verbal Learning and Verbal Behavior, 1(3):79-84. find
  5. Deese, James E. (1964). "The associative structure of some common English adjectives." Journal of Verbal Learning and Verbal Behavior, 3(5):347-57. find
  6. Deese, James E. (1965). The Structure of Associations in Language and Thought. The Johns Hopkins Press. find
  7. Ervin, Susan (1961). "Changes with age in the verbal determinants of word-association." American Journal of Psychology 74:361-72. find
  8. Ervin, Susan (1963). "Correlates of associative frequency." Journal of Verbal Learning and Verbal Behavior, 1(6):422-31. find
  9. Fenk-Oczlon, Gertraud (1989). "Word frequency and word order in freezes." Linguistics, 27(3):517-56. find
  10. Francis, Winthrop N. and Ku~era, Henry (1982). Frequency Analysis of English Usage: Lexicon and Grammar. Houghton Mifflin. find
  11. Gross, Derek; Fischer, Ute; and Miller, George A. (1989). "Antonymy and the representation of adjectival meanings." Journal of Memory and Language, 28(1):92-106. find
  12. Shanks, David R. and Dickinson, A. (1987). "Associative accounts of causality judgment." In The Psychology of Learning and Motivation, edited by Gordon H. Bower, 21:229-61, Academic Press. find
  13. Thorndike, Edward L. and Lorge, Irving (1944). The Teacher's Word-Book of 30,000 Words. Teacher's College, Columbia University. find
  14. Wilks, Yorick; Fass, Dan; Guo, Cheng-Ming; McDonald, James E.; Plate, Tony; and Slater, Brian M. (1989). "Machine tractable dictionaries as tools and resources for natural language processing." In Computational Lexicography for Natural Language Processing, edited by Branimir Boguraev and Edward J. B… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/justeson-1991-co-occurrences-antonymous-adjectives-their/