Can BERT predict fillers for construction elements?
International Journal of Corpus Linguistics · 19.06.2026 · 31 (2) · 228-260
DOI: 10.1075/ijcl.24108.fel ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Corpus Linguistics, online erschienen am 21.05.2026. Verfasst von Tim Feldmüller und 2 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben.
Abstract (Originalsprache)
The starting point of this paper is a central problem in constructicography: on the one hand, constructicon projects aim at describing a broad spectrum of constructions based on usage data; on the other hand, the established method for determining salient fillers of construction elements (slots) requires comprehensive annotations of entire corpora. In this paper, we propose a new method to tackle this problem. Using the pre-trained language model BERT and a limited number of example sentences, the method predicts typical fillers of slots, so-called collo-profiles. A critical evaluation of the method reveals not only its potentials and limitations but also differences between such a prediction-based method and classical count-based methods.
Zitieren
APA
Feldmüller, T., Barteld, F., & Ziem, A. (2026). Can BERT predict fillers for construction elements?. International Journal of Corpus Linguistics, 31(2), 228-260. https://doi.org/10.1075/ijcl.24108.fel
Vancouver
Feldmüller T, Barteld F, Ziem A. Can BERT predict fillers for construction elements?. International Journal of Corpus Linguistics. 2026;31(2):228-260. doi:10.1075/ijcl.24108.fel
BibTeX
@article{Feldmller2026,
author = {Feldmüller, Tim and Barteld, Fabian and Ziem, Alexander},
title = {Can BERT predict fillers for construction elements?},
journal = {International Journal of Corpus Linguistics},
year = {2026},
volume = {31},
number = {2},
pages = {228--260},
doi = {10.1075/ijcl.24108.fel}
}
EUVIMED