BOOKVERSE: A Full-Stack Web-Based Book Marketplace with Collaborative Filtering Recommendation System

Abstract

The rapid growth of e-commerce platforms has transformed how users discover and purchase products online. However, traditional online bookstores often lack intelligent mechanisms for personalized recommendations. This paper presents BOOKVERSE, a full-stack web-based book marketplace integrating user-based collaborative filtering with real-time user interaction data. The system enables users to browse, search, and purchase books while generating personalized recommendations based on behavioral data such as browsing history and cart activity. A cosine similarity-based approach is used to identify similar users and recommend relevant books. The platform also includes a rule-based chatbot for assistance and an admin analytics dashboard for monitoring system performance. Experimental evaluation demonstrates a precision of 0.72 and an average response time of 0.35 seconds, indicating satisfactory performance. The proposed system enhances user engagement and improves the efficiency of book discovery.