Development Of A Book Recommendation System Using Collaborative Filtering

Abstract

In the rapidly evolving digital era, the need for accurate and personalized recommendation systems is increasingly important, particularly in digital libraries and online bookstores. This study aims to develop a book recommendation system using a collaborative filtering approach, which leverages user interaction data to suggest books that align with individual preferences. The system utilizes a user-based collaborative filtering method by calculating similarities between users based on their historical book ratings. The dataset used in this research is a simulated, anonymized dataset from a school library. Testing results indicate that the system is capable of delivering relevant recommendations with good accuracy, demonstrated by a low Mean Absolute Error (MAE) score and positive user feedback. This system allows users to discover books aligned with their interests more efficiently, thereby enhancing the overall reading experience.