prof. dr. ir. Thomas Demeester
- Work address
-
Technologiepark Zwijnaarde 126
9052 Zwijnaarde - Thomas.Demeester@UGent.be
- ORCID iD
-
0000-0002-9901-5768
Show
Sort by
-
- Conference Paper
- open access
Detecting Idiomatic Multiword Expressions in Clinical Terminology using Definition-Based Representation Learning
-
- Journal Article
- A1
- open access
EduQG : a multi-format multiple-choice dataset for the educational domain
-
- Journal Article
- A1
- open access
Learning to reuse distractors to support multiple choice question generation in education
-
CookDial : a dataset for task-oriented dialogs grounded in procedural documents
-
DLIME-graphs : a DLIME extension based on triple embedding for graphs
-
- Conference Paper
- open access
BioLORD: Learning ontological representations from definitions for biomedical concepts and their textual descriptions
(2022) p.1454-1465 -
Next-year bankruptcy prediction from textual data: Benchmark and baselines
(2022) p.187-195 -
- Conference Paper
- C1
- open access
Neural Bayesian network understudy
-
- Conference Paper
- C3
- open access
Visual conversation starters for human-robot interaction
-
Robustifying sentiment classification by maximally exploiting few counterfactuals