W
Walter Daelemans
Researcher at University of Antwerp
Publications - 463
Citations - 13831
Walter Daelemans is an academic researcher from University of Antwerp. The author has contributed to research in topics: Language technology & Natural language. The author has an hindex of 57, co-authored 444 publications receiving 12732 citations. Previous affiliations of Walter Daelemans include VU University Amsterdam & Radboud University Nijmegen.
Papers
More filters
Using syntactic features to predict author personality from text
Kim Luyckx,Walter Daelemans +1 more
TL;DR: A successful methodology takes a two-stage approach which achieves automatic selection of features with high predictive value for the categories to be learned, and uses machine learning algorithms to learn to categorize new documents by using the selected features.
Posted Content
Bootstrapping a Tagged Corpus through Combination of Existing Heterogeneous Taggers
Jakub Zavrel,Walter Daelemans +1 more
TL;DR: Experiments show that COMBI-BOOTSTRAP can integrate a wide variety of existing resources, and achieves much higher accuracy than both the best single tagger and an ensemble tagger constructed out of the same small training sample.
Journal ArticleDOI
Data Mining as a Method for Linguistic Analysis: Dutch Diminutives
TL;DR: In this article, the authors propose to use data mining techniques (inductive techniques for the automatic acquisition of comprehensibl e knowledge from data) as a method in linguistic analysis, which can assist in linguistic theory formation by providing a new tool for the evaluation of linguistic hypotheses, for the extraction of rules from corpora, and for the discovery of useful linguistic categories.
Journal ArticleDOI
Lemmatization for variation-rich languages using deep learning
TL;DR: A novel approach to sequence tagging for languages that are rich in (e.g. orthographic) surface variation, based on recent advances in the field of ‘deep’ representation learning, where neural networks have led to a dramatic increase in performance across several domains.