Linguists map Mapudungun verb patterns to improve language processing tool for indigenous language

Researchers studying Mapudungun, an indigenous language spoken in Chile and Argentina, have systematically classified how verbs change form by examining which word endings can attach to which verb roots — the basic building blocks of the language. This work feeds directly into Dungupeyum, a computational tool designed to analyze Mapudungun text automatically, making the tool more accurate at recognizing and parsing verb forms.

Mapudungun is spoken by roughly 100,000–200,000 people and has historically received minimal computational linguistics investment compared to major languages. Building better language analysis tools is foundational infrastructure for digital literacy, education, and cultural preservation — without it, indigenous languages fall further behind in the digital economy, from search engines to translation to AI training data. This work is unglamorous but necessary: someone has to do the grinding morphological taxonomy work before you can build anything that actually works.