Exploring and Implementing Estonian Text Simplification Using Machine Learning
Name
Sten Marcus Malva
Abstract
The goal of this bachelor’s thesis was to explore Estonian text simplification using machine learning and to train such a model with Estonian data. In order to train it, a significant amount of data was gathered, translated, corrected and generated. Then previously implemented algorithms for English were evaluated and one of them was used to take as a base for the Estonian model. Finally a model was trained in three different configurations and evaluated.
Graduation Thesis language
English
Graduation Thesis type
Bachelor - Computer Science
Supervisor(s)
Eduard Barbu
Defence year
2024