Performance Evaluation of Sentiment Analysis on Reddit Comments: Insights and Improvement Opportunities for Naive Bayes, SVM, and BERT Models

Sentiment analysis plays a key role in Natural Language Processing (NLP). It aims to pull out sentiments, opinions, and emotions from text. Social platforms like Reddit show how important this is. People express many different emotions there. This makes sentiment analysis useful for market analysis, consumer feedback, and social behavior research. This study tackles problems in emotion classification. These include not having enough emotion databases, making things too simple with binary classification, and the high computing costs of processing big datasets like GoEmotions. We use the GoEmotions dataset to test machine learning and transformer-based models. This dataset has 58,000 comments labeled with 27 emotions. We look at Support Vector Machines (SVM), Naive Bayes, and BERT. Our findings show that BERT does better than the others. It’s more accurate and captures small differences in sentiment well. BERT’s strong performance shows it’s a good fit for complex sentiment analysis tasks. This paper stresses the need for accuracy and efficient computing. It sets a standard to help push fine-grained sentiment analysis research forward.