Associative and Semantic Features Extracted From Web-Harvested Corpora
Elias Iosif, Maria Giannoudaki, Eric Fosler-Lussier, Alexandros Potamianos
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%.- Anthology ID:
- L12-1301
- Volume:
- Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)
- Month:
- May
- Year:
- 2012
- Address:
- Istanbul, Turkey
- Venue:
- LREC
- SIG:
- Publisher:
- European Language Resources Association (ELRA)
- Note:
- Pages:
- 2991–2998
- Language:
- URL:
- http://www.lrec-conf.org/proceedings/lrec2012/pdf/536_Paper.pdf
- DOI:
- Cite (ACL):
- Elias Iosif, Maria Giannoudaki, Eric Fosler-Lussier, and Alexandros Potamianos. 2012. Associative and Semantic Features Extracted From Web-Harvested Corpora. In Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12), pages 2991–2998, Istanbul, Turkey. European Language Resources Association (ELRA).
- Cite (Informal):
- Associative and Semantic Features Extracted From Web-Harvested Corpora (Iosif et al., LREC 2012)
- PDF:
- http://www.lrec-conf.org/proceedings/lrec2012/pdf/536_Paper.pdf