<p>E-commerce growth has increased rapidly in recent years, and several people utilize this popular channel to purchase services and products through Internet resources. Shopping sites have become very important for customers to buy the best products, and sales have increased for this resource. Furthermore, people purchasing through internet resources face several problems and a lot of confusion due to the massive number of products. They find it very difficult to choose their favourite product. In the present market, various popular traditional algorithms named Collaborative Filtering (CF), Planned Behavioral Theory (PBT), Markov Hidden Model (MHM), Traditional Machine Learning (TMC) and Analysis of Component (AoC) were used in the E-commerce sites for the users to purchase and choose the products in a customized manner with high service and more loyalty. In the traditional methods, Customers face several difficulties in determining the statistical probability of the product due to bulk data information during shopping. In this research, the heuristic computational method (FHCM) has been integrated with e-commerce sites, which helps optimize product search during the purchase and customer authentication process in an effective manner. This proposed method has been experimentally analyzed at lab scale testbed software and found to be more helpful in solving the problems in data sparse to identify the best product on the site for the customers.</p>