E-commerce has recently experienced a remarkable development. As a result, more people are making purchases online, which has increased the number of product reviews posted online. It has also revolutionised the way we shop, making products easily accessible to customers without leaving the comfort of their homes. As a result, the importance of product reviews has become increasingly significant. With countless reviews available online, it can be a daunting task for customers to navigate through them all to make an informed decision. Evaluation of text data and the extraction of the sentiment component frequently use the field of sentiment analysis. Textual information is produced every day by client reviews, comments, communications, and suggestions on online business websites. However, with the advancements in machine learning, sentiment analysis has become a powerful tool to simplify this process. In this paper, we investigate how to polarise reviews with acceptable accuracy using supervised learning on a sizable Amazon dataset.