Internet users are a huge segment of the consumer market, and businesses nowadays are trying to enter e-commerce, where customers leave reviews regarding products and services. Sentiment analysis is the process of extracting the customer's real feelings from the reviews of the product or services. This study compares logistic regression, naive Bayes, neural networks, and support vector machine algorithms for sentiment analysis and finds the best-performing classifiers among them. This applied study evaluates the classifiers using accuracy, precision, recall, and F1-score metrics. The dataset was taken from the E-Commence website, on which NLP and other classifiers are employed. The results show that the Naive Bayes model, with 94% accuracy, outperforms the different classifiers, where Logistic Regression and Neural Networks are at a similar level of 93%. In comparison, the SVM gave us an average of about 92%. This study suggests the significance of continuously updating sentiment analysis systems to maintain accuracy and relevance. Real-time sentiment analysis tools are a good technique for any text mining work that can help companies address customer problems based on immediate feedback and improve their products.