This study analyzes Amazon product reviews using machine learning, deep learning, and transformer-based techniques to classify sentiments into positive, negative, or neutral. The research uses a dataset of 400,000 Amazon product reviews across five categories: mobile electronics, furniture, camera, grocery, and watches. The dataset was preprocessed to handle missing values, convert text to lowercase, remove stop words, and tokenize the text. Feature extraction methods included Bag-of-Words (BOW) and Term Frequency-Inverse Document Frequency (TF-IDF). The study evaluated various machine learning algorithms, including Multinomial Naive Bayes (MNB), Random Forest (RF), Decision Tree (DT), and Logistic Regression (LR), as well as ensemble learning techniques like Bagging. Deep learning models such as Convolutional Neural Networks (CNNs), Bidirectional Long Short-Term Memory (Bi-LSTM), and transformer-based models like BERT and XLNet were also tested. The results showed that BERT achieved the highest accuracy of 89%, outperforming other models. The study also explored the use of the Bi-LSTM model, which achieved an accuracy of 87%. The BERT model was further analyzed using the Local Interpretable Model-Agnostic Explanations (LIME) method to understand its predictions. The BERT model demonstrated strong performance in classifying sentiments, with high accuracy and F1 scores. The research compared the proposed approach with state-of-the-art methods and found that the BERT model outperformed existing models on various datasets. The study also identified limitations, such as the difficulty in distinguishing between similar sentiments like 4-star and 5-star ratings, and 1-star and 2-star ratings. Future work could involve expanding the analysis to include reviews in different languages and cultural contexts, as well as incorporating user metadata to improve sentiment classification accuracy. The study highlights the effectiveness of transformer-based models in sentiment analysis and their potential for real-world applications.This study analyzes Amazon product reviews using machine learning, deep learning, and transformer-based techniques to classify sentiments into positive, negative, or neutral. The research uses a dataset of 400,000 Amazon product reviews across five categories: mobile electronics, furniture, camera, grocery, and watches. The dataset was preprocessed to handle missing values, convert text to lowercase, remove stop words, and tokenize the text. Feature extraction methods included Bag-of-Words (BOW) and Term Frequency-Inverse Document Frequency (TF-IDF). The study evaluated various machine learning algorithms, including Multinomial Naive Bayes (MNB), Random Forest (RF), Decision Tree (DT), and Logistic Regression (LR), as well as ensemble learning techniques like Bagging. Deep learning models such as Convolutional Neural Networks (CNNs), Bidirectional Long Short-Term Memory (Bi-LSTM), and transformer-based models like BERT and XLNet were also tested. The results showed that BERT achieved the highest accuracy of 89%, outperforming other models. The study also explored the use of the Bi-LSTM model, which achieved an accuracy of 87%. The BERT model was further analyzed using the Local Interpretable Model-Agnostic Explanations (LIME) method to understand its predictions. The BERT model demonstrated strong performance in classifying sentiments, with high accuracy and F1 scores. The research compared the proposed approach with state-of-the-art methods and found that the BERT model outperformed existing models on various datasets. The study also identified limitations, such as the difficulty in distinguishing between similar sentiments like 4-star and 5-star ratings, and 1-star and 2-star ratings. Future work could involve expanding the analysis to include reviews in different languages and cultural contexts, as well as incorporating user metadata to improve sentiment classification accuracy. The study highlights the effectiveness of transformer-based models in sentiment analysis and their potential for real-world applications.