Enhancing Book Recommendation Systems Using Collaborative Approach: A Comprehensive Study

In this study, we delve into the creation of a Book Recommendation System utilizing collaborative filtering techniques to offer customized book suggestions to users. Beginning with the cleansing of a dataset encompassing book details, user information, and their ratings, we ensure data accuracy for our analysis. Within this dataset, we uncover intriguing insights such as user demographics, popular authors, and book ratings, which inform our recommendation framework. Leveraging collaborative filtering, particularly Singular Value Decomposition, in order to separate the user-item matrix to generate tailored recommendations based on users' historical interactions. To gauge the system's effectiveness, we assess its performance using metrics such as recall@5 and recall@10, demonstrating its capability to deliver pertinent book recommendations. Additionally, we explore alternative recommendation strategies including popularity-based and author-based approaches, providing a comprehensive comparative analysis. Ultimately, our Book Recommendation System aims to enhance user book selection experiences, fostering greater engagement and satisfaction within the reading community.