E-COMMERCE PRODUCT RECOMMENDATION SYSTEM USING COLLABORATIVE FILTERING

1R Uma, 2 P Dhanush Kumar, 3 Samala Akhil, 4 J Sai Charan,5 K Jaipal Reddy

doi:10.62643/

Authors

1R Uma, 2 P Dhanush Kumar, 3 Samala Akhil, 4 J Sai Charan,5 K Jaipal Reddy Author

DOI:

https://doi.org/10.62643/

Keywords:

E-Commerce Recommendation System, Collaborative Filtering, User–Item Interaction Matrix ,Cosine Similarity ,Personalized Product Recommendation ,Machine Learning in E-Commerce

Abstract

The rapid growth of e-commerce platforms has created challenges for users in identifying relevant products among a large number of available options. This project proposes an E-Commerce Product Recommendation System using Collaborative Filtering to provide personalized product suggestions based on user preferences and historical rating data. The system utilizes an Amazon product review dataset containing over two million transaction records with attributes such as User ID, Product ID, ratings, and timestamps. After data preprocessing, a User–Product Interaction Matrix is constructed, and Cosine Similarity is applied to identify users with similar rating patterns. Based on the preferences of these similar users, the system generates Top-N product recommendations for a target user. The experimental results demonstrate that the model successfully identifies similar users and recommends products with high predicted ratings, improving recommendation relevance. The system effectively demonstrates how collaborative filtering techniques can enhance personalized shopping experiences and assist e-commerce platforms in improving customer engagement and product discovery.