I wrote a popular science blog (in Dutch), about my PhD research, on the Brussels science popularization platform Wtnschp.be, which has subsequently also been published on the website of Belgian science magazine Eos Wetenschap. I explain morphological simplification, specifically in the language Alorese in Eastern Indonesia, and how agent-based models can be used to study this.
How can AI techniques be used to help language preservation and documentation efforts? Together with my colleague Katie Mudd, I will organize a half-day virtual event, which will provide an introduction to how agent-based computer simulations can support field linguists and development organizations. Agent-based models can be used to simulate the factors that endanger languages, and evaluate possible measures that can be taken before implementing policies in real life.
The workshop will take place virtually on Thursday 25 March 2021, 14:00-17:00h CET (Brussels). More information and registration can be found on the event webpage.
Ronald Veldhuizen wrote an article in De Volkskrant (Dutch newspaper) Gaat een steenoud sterrenbeeldverhaal helemaal terug naar de prehistorie?, about a theory by Julien d’Huy on the reconstruction of an ancestral tree of myths. I gave a short comment on the theory in the article, based on my experiences with reconstructing phylogenetic trees of languages. Jamie Tehrani also gave his vision and compared the theory to his own models.
A pdf of the article can be found here.
The paper “Word prediction in computational historical linguistics”, with Willem Zuidema, has been published in the Journal of Language Modelling. The paper is based on my master thesis research, on supporting historical linguistics by predicting words between languages. The article is available here.