Abstract

Finite-state morphological modeling has been used in natural language processing for many years particularly when dealing with lower resource languages. The present study details the development of an open-source finite-state morphological model for the Q'eqchi' Maya language using Helsinki Finite-State Technology (HFST) and the Giellatekno infrastructure. This project represents the first comprehensive morphological analyzer for Q'eqchi' and sets a foundation for future work in data annotation for this language. The resulting transducer consists of 4,439 lexemes, 2,610 states and 9,558 transitions and covers between 75% and 85% of tokens in a Q'eqchi' corpus. The success of this project lays the groundwork for future work in improved automatic corpus annotation in Q'eqchi', as well as suggesting further success in the development of similar utilities for other Mayan languages.

Degree

MA

College and Department

Humanities; Linguistics

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2023-12-08

Document Type

Thesis

Handle

http://hdl.lib.byu.edu/1877/etd13044

Keywords

Mayan languages, finite-state morphology, Q'eqchi', HFST, Giellatekno

Language

english

Download

Share

COinS