Sirvi Autor "Aruoja, Hendrik" järgi

Nüüd näidatakse 1 - 1 1

Automaatmärgendatud süntaksi ja morfoloogia konfliktide lahendamine
(Tartu Ülikool, 2025) Aruoja, Hendrik; Laur, Sven, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Arvutiteaduse instituut
The Bachelor's thesis focuses on finding cost-effective ways to resolve syntax-morphology conflicts for EstNLTK that utilizes instructing of Large Language Models (LLMs) as an alternative to manually labeling. The goal is to create a Python scripts that includes database query generation, execution of queries and processing of LLM prompts via API. The work is divided into a theoretical part, where the principles of syntax and morphology analysis are introduced, and a practical part, where the created python scripts and their capabilities are described.