This research paper intends to explore the methodologies and effectiveness of artificial intelligence (AI) in classifying different emotions through sentiment analysis. Sentiment analysis, also known as opinion mining, utilizes natural language processing (NLP), computational linguistics, and text analysis to systematically identify, extract, quantify, and study affective states and subjective information. The study provides an overview of the current landscape of sentiment analysis, highlighting widely used techniques such as traditional machine-learning models, advanced deep-learning models such as Long Short-Term Memory (LSTM) and Transformer (E.g., BERT, GPT) models, as well as their shortcomings with current standards. A detailed methodology is presented, focusing on data collection, preprocessing, model training, and evaluation. The LSTM model created for this research demonstrates high performance in capturing long-term dependencies in text, marking around an 85% accuracy rate. The results underscore the significant advancements in AI-driven sentiment analysis and its applications across various industries, including marketing, customer service, and market research. Furthermore, the study highlights the potential of sentiment analysis in the mental health domain, where it can facilitate early detection and intervention for mental health issues through the analysis of textual data. The findings contribute to the ongoing development and refinement of AI techniques for more accurate and nuanced emotion classification.
Krish Kalla (Sat,) studied this question.