<![CDATA[Advancements in Large Language Models (LLMs) have opened new opportunities for emotion analysis in social media comments. This study aims to explore the application of LLMs in classifying users' emotions based on their comment texts across various social media platforms. The methodology involves collecting comment data from Twitter, Facebook, and YouTube, followed by text preprocessing using Natural Language Processing (NLP) techniques. LLMs such as GPT and BERT are evaluated for their ability to detect primary emotions including happiness, sadness, anger, and fear. The results show a precision of 0.89 (89%), recall of 0.80 (80%), and an F1-score of 0.84 (84%). These findings indicate that LLMs offer higher accuracy compared to conventional sentiment analysis methods, particularly in their ability to understand linguistic context and nuances.]]>
Copyrights © 2025
