QLVL members - Thomas Wielfaert
Thomas holds a Master's of English literature and linguistics from Ghent University (2009) and a Master's of Advanced Studies in Linguistics: interdisciplinary linguitics (major: computational linguistics) from the University of Antwerp (2010). He is pursuing a PhD on distributional semantics (so-called Word Sense Induction) for lexical semantic purposes under the supervision of prof. dr. Dirk Speelman and prof. dr. Dirk Geeraerts.
E-mail: thomas.wielfaert@kuleuven.be
Phone: +32 16 37 24 69
Surface mail: Research Unit Linguistics, Blijde-Inkomststraat 21 PO
Box 03308, B-3000 Leuven, Belgium
Office: room 02.37 of the Faculty of Arts of the University of Leuven
Research
Thomas's research is part of the 3H120217 OT-project "From lexical to semantic sociolectometry: New methods for the corpus-based analysis of variation in lexical categorization." funded by the KU Leuven Research Fund (BOF). Token-level Semantic Vector Space models (an extension of Hinrich Schütze's 1998 paper) are used to model semasiological variation, i.e. polysemy, in an automatic and unsupervised way.
Other research interests include:
- (Linguistic) data visualisation
- Language variation and change, with a focus on Dutch
- Topic modelling (LDA) for semantic purposes
- Applications of distributional semantics for language technology
- Cluster analysis techniques
Representative publications
Heylen, K., T. Wielfaert, D. Geeraerts & D. Speelman (2015) "Monitoring Polysemy: Word Space Models as a Tool for Large-Scale Lexical Semantic Analysis." Lingua: International Review of General Linguistics, 157, 153-172.
Wielfaert, T., K. Heylen & D. Speelman (2013). "Visualisations interactives des espaces vectoriels sémantiques pour l'analyse lexicologique." Actes de SemDis 2013 : Enjeux actuels de la sémantique distributionnelle., Traitement Automatique du Langage Naturel. Les Sables-d'Olonne, 17-21 June 2013 (pp. 154-166). Les Sables d'Olonne, France.
