| Title |
Sentiment analysis of Lithuanian online reviews using large language models / |
| Authors |
Vileikyte, Brigita ; Lukoševičius, Mantas ; Stankevičius, Lukas |
| DOI |
10.15388/Proceedings.2024.44 |
| Full Text |
|
| Is Part of |
IVUS2024: 29th international conference "Information society and university studies", Vilnius University, Kaunas Faculty, Kaunas, Lithuania, May 17th, 2024: abstracts.. Vilnius : Vilniaus universiteto leidykla. 2024, p. 34 |
| Abstract [eng] |
Sentiment analysis is a widely researched area within Natural Language Processing, attracting significant interest from various companies and resellers online due to the advent of automated solutions. Despite this, the task remains challenging because of the inherent complexity of languages and the subjective nature of sentiments. In this paper, we strive to evaluate texts as objectively as possible, focusing on Lithuanian online reviews from multiple domains. We discuss the outcomes of a sentiment analysis performed with a relatively small dataset in this less commonly studied language. Our review of existing Lithuanian NLP research reveals that traditional machine-learning methods and classification algorithms have limited effectiveness in analyzing and detecting sentiments under resource constraints. Additionally, this work explores the capabilities of pre-trained multilingual Large Language Models, specifically detailing our experiences with fine-tuning BERT and T5 models for sentiment analysis on a multitopic Lithuanian review dataset. |
| Published |
Vilnius : Vilniaus universiteto leidykla |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2024 |
| CC license |
|
