Authors
Paul Van Eecke, Katrien Beuls
Published in
iScience. Volume 29. Issue 11. Pages 118129. Nov 20, 2026. Epub Nov 05, 2026.
Abstract
This article introduces PyFCG, an open-source scientific software toolkit that enables its users to efficiently integrate state-of-the-art constructionist language processing and learning into larger Python projects. The toolkit provides a fully Pythonic interface to the leading computational construction grammar framework, Fluid Construction Grammar (FCG), facilitating its swift integration with Python's rich ecosystem of libraries, in particular those implementing today's most popular machine learning techniques. Apart from a general description of the toolkit, this article provides three illustrative case studies that demonstrate how PyFCG facilitates (i) the formalization and computational operationalization of construction grammar analyses, (ii) the learning of usage-based construction grammars from corpora of language use, and (iii) the implementation of multi-agent emergent communication experiments that combine symbolic and neural computation.
PMID:
43034531
Bibliographic data and abstract were imported from PubMed on 15 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 5
- Comments 0