E-commerce is getting out of control—there are so many options, it’s hard to find the best deals. Regular price comparison websites don’t really help, they’re too basic. This paper tries to fix that by using machine learning. We combined a few different techniques to create a personalised recommendation system. We tested it out and the results were pretty good—people were more satisfied, engaged, and actually bought things. This is just the beginning of using machine learning in e-commerce. This research looks at how machine learning methods may be used to enhance the price comparison process. The research emphasises the need of gathering a range of reliable datasets, using effective preprocessing methods, and establishing mechanisms for real-time updates. The developed machine learning models make use of product features and price information to provide precise and customised recommendations via an intuitive user interface. Utilising HTML, CSS, JavaScript, Flask, and Python, the proposed solution creates an easy-to-use online application that empowers customers to make informed purchase choices. The system uses the predictive abilities of linear regression to accurately estimate product price based on historical data and relevant attributes, facilitating seamless comparisons across various e-commerce platforms. This innovative approach aims to enhance the online shopping experience while providing useful data for future developments in the e-commerce industry.