SENTIMENT ANALYSIS OF PHONE SALES USING MACHINE LEARNING

This study focuses on analysing customer sentiments in mobile phone reviews using machine learning and deep learning techniques. A dataset of 4,500 Amazon phone reviews was processed using natural language processing methods such as tokenization, lemmatization, and stop-word removal. TF-IDF was used for machine learning models, while embeddings supported deep learning. Logistic Regression, Random Forest, and LSTM models were implemented and evaluated using accuracy, precision, recall, and F1-score. Results showed that LSTM achieved the highest accuracy (96.5%), outperforming Logistic Regression (83.09%) and Random Forest (80.88%). Despite challenges such as class imbalance, the models demonstrated effectiveness in sentiment classification, providing valuable insights for e-commerce decision-making.