Associative and semantic features extracted fromweb-harvested corpora

Elias Iosif, Maria Giannoudaki, Eric Fosler-Lussier, Alexandros Potamianos

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We address the problem of automatic classification of associative and semantic relations between words, and particularly those that hold between nouns. Lexical relations such as synonymy, hypernymy/hyponymy, constitute the fundamental types of semantic relations. Associative relations are harder to define, since they include a long list of diverse relations, e.g., "Cause-Effect", "Instrument-Agency". Motivated by findings from the literature of psycholinguistics and corpus linguistics, we propose features that take advantage of general linguistic properties. For evaluation we merged three datasets assembled and validated by cognitive scientists. A proposed priming coefficient that measures the degree of asymmetry in the order of appearance of the words in text achieves the best classification results, followed by context-based similarity metrics. The web-based features achieve classification accuracy that exceeds 85%.

Original languageEnglish
Title of host publicationProceedings of the 8th International Conference on Language Resources and Evaluation, LREC 2012
EditorsMehmet Ugur Dogan, Joseph Mariani, Asuncion Moreno, Sara Goggi, Khalid Choukri, Nicoletta Calzolari, Jan Odijk, Thierry Declerck, Bente Maegaard, Stelios Piperidis, Helene Mazo, Olivier Hamon
PublisherEuropean Language Resources Association (ELRA)
Pages2991-2998
Number of pages8
ISBN (Electronic)9782951740877
Publication statusPublished - 2012
Externally publishedYes
Event8th International Conference on Language Resources and Evaluation, LREC 2012 - Istanbul, Turkey
Duration: 21 May 201227 May 2012

Publication series

NameProceedings of the 8th International Conference on Language Resources and Evaluation, LREC 2012

Conference

Conference8th International Conference on Language Resources and Evaluation, LREC 2012
Country/TerritoryTurkey
CityIstanbul
Period21/05/1227/05/12

Keywords

  • Associative relations
  • Priming
  • Semantic relations

Fingerprint

Dive into the research topics of 'Associative and semantic features extracted fromweb-harvested corpora'. Together they form a unique fingerprint.

Cite this