Analysing user sentiments in social media: the supremacy of deep learning methods over traditional machine learning techniques

Sentiment analysis plays a crucial role in natural language processing by transforming large-scale textual data from social media into meaningful business insights. This study explores three key sentiment analysis approaches; emotion detection sentiment analysis, aspect-based sentiment analysis, and intent-based sentiment analysis and evaluates the performance of traditional machine learning and deep learning models. The study assesses machine learning algorithms such as support vector machine, multinomial Naïve Bayes, logistic regression, decision tree, and random forest against deep learning models, including BERT, LSTM, CNN, RNN, and GRU. Performance is evaluated using accuracy, precision, recall, and F1-score. The findings indicate that deep learning models outperform traditional machine learning techniques, with LSTM achieving the highest accuracy of 93.91%. The study provides valuable insights into the effectiveness of different sentiment analysis models, highlighting deep learning's advantage in capturing complex sentiment structures.