Main Article Content
The world that we see today is becoming increasingly digital. By making products accessible to customers without the need for them to leave their homes, e-commerce is gaining ground in this digitalized world. As more and more people rely on online products, reviews are becoming increasingly important. A customer must read thousands of reviews to understand a product before making a purchase. However, utilizing a model to polarize those reviews and learn from them would make going through thousands of reviews much simpler in this prosperous age of deep learning. On a massive Amazon dataset, we polarized it using deep learning techniques and achieved satisfactory accuracy. Reviews on Amazon aren't just about the product—they're also about the customer service. In this paper, we propose a system that categorizes customer reviews and then determines the sentiment of those reviews. This will make it easier for customers to make decisions if they can clearly distinguish between product reviews and service reviews. Product feature sentiment is also extracted using a rule. Additionally, we provide a visualization for our summary of the results.