A uniform approach to analogies, synonyms, antonyms, and associations

Abstract

/npsi/ctrl?action=rtdoc&an=5764174&lang=en http://nparc.cisti-icist.nrc-cnrc.gc.ca/npsi/ctrl?action=rtdoc&an=5764174&lang=frAccess and use of this website and the material on it are subject to the Terms and Conditions set forth at http://nparc.cisti-icist.nrc-cnrc.gc.ca/npsi/jsp/nparc_cp.jsp?lang=en (Coling 2008(Coling ), pp. 905-912, 2008 A Uniform Approach to Analogies, Synonyms, Antonyms, and Associations Turney, Peter D. NRC Publications Archive Archives des publications du CNRC Proceedings of the 22nd International Conference on Computational Linguistics AbstractRecognizing analogies, synonyms, antonyms, and associations appear to be four distinct tasks, requiring distinct NLP algorithms. In the past, the four tasks have been treated independently, using a wide variety of algorithms. These four semantic classes, however, are a tiny sample of the full range of semantic phenomena, and we cannot afford to create ad hoc algorithms for each semantic phenomenon; we need to seek a unified approach. We propose to subsume a broad range of phenomena under analogies. To limit the scope of this paper, we restrict our attention to the subsumption of synonyms, antonyms, and associations. We introduce a supervised corpus-based machine learning algorithm for classifying analogous word pairs, and we show that it can solve multiple-choice SAT analogy questions, TOEFL synonym questions, ESL synonym-antonym questions, and similar-associated-both questions from cognitive psychology.

Le test de substituabilité à l'épreuve des corpus : utiliser l'analyse distributionnelle automatique pour l'étude des relations lexicales

“…On peut ainsi citer les travaux de Grefenstette (1992) portant sur des textes issus du domaine médical, ou, parmi les études réalisées à partir de textes en français sur le même domaine, les travaux de Nazarenko et al (1997) et Bouaud et al (2000). L'accroissement de la quantité de textes accessibles au format électronique a permis par la suite au modèle distributionnel de gagner en popularité en entrainant l'extension de son domaine d'application aux textes non spécialisés comme les textes journalistiques (Dias et al 2010) ou le Web (Turney 2008 ;Agirre et al 2009). Le fait d'appliquer l'analyse distributionnelle (AD) à des textes ne relevant pas de domaines de spécialité produit toutefois des résultats moins faciles à caractériser.…”

“…Plusieurs travaux ont montré néanmoins que ces techniques pouvaient permettre de repérer des relations sémantiques de différents types -synonymie, généricité, analogie, etc. (Turney 2008 ;Baroni et Lenci 2011).…”

Ontological Lexicon Enrichment: The Badea System For Semi-Automated Extraction Of Antonymy Relations From Arabic Language Corpora

“…A study that employs a supervised approach is presented in Turney [35]. It uses a machine learning algorithm for pattern identification that classifies analogous (synonyms, antonyms, and associations) word pairs and can be used to solve multiple-choice analogy, synonym, and synonym-antonym questions.…”

Extraction of Hyponymy, Meronymy and Antonymy Relation Pairs : A Brief Survey

“…Turney [18] classified synonyms, antonyms, and semantically similar words by supervised machine learning methods. Previously obtained patterns were used as feature vectors and cooccurrence frequency of pairs with these patterns in 5x10 10 word corpus was used as feature values as well as.…”