Abstractive summarization of news in  Ukrainian

dc.contributor.author	Vozniak, Yuliia
dc.date.accessioned	2024-08-26T11:53:39Z
dc.date.available	2024-08-26T11:53:39Z
dc.date.issued	2024
dc.identifier.citation	Vozniak Yuliia. Abstractive summarization of news in Ukrainian.Ukrainian Catholic University, Faculty of Applied Sciences, Department of Computer Sciences. Lviv 2024, 38 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4682
dc.language.iso	en	uk
dc.subject	news	uk
dc.subject	news in Ukrainian	uk
dc.subject	Abstractive summarization	uk
dc.title	Abstractive summarization of news in Ukrainian	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	Significant advancements in large language models in recent years have brought rapid improvements in all areas of text processing, including text summarization. However, in the case of low-resource languages like Ukrainian, this improvement is underresearched. In this thesis we take a snapshot of the current state of abstrac- tive text summarization of Ukrainian news. To achieve this, we review the datasets available for the Ukrainian language, highlight existing problems and propose a possible improvement by introducing a custom news summarization dataset cre- ated by matching summaries from Telegram channels with their respective articles on the news media websites. We also experiment with text summarization via LLM Llama-3 and analyze the results both quantitatively and qualitatively.	uk