Bookgram: Book exchange system with a Hybrid recommendation engine.

Abstract

"In an age dominated by digital media, the enduring appeal of printed books persists due to the unique tactile experience they offer. However, traditional bookstores often suffer from limited space, hindering the discovery of new books matching one's interests. Moreover, the high cost of new books can be a barrier to accessing literature and educational materials. To address these challenges and foster sustainability, wider access to printed books, and reading engagement, a book exchange system with a hybrid recommendation engine was developed.

The system features a user-friendly interface, facilitating book exchanges, with capabilities for book management, search, and messaging. Additionally, a recommendation engine was incorporated, utilizing a Long Short-Term Memory model to predict users' book genre preferences based on their input. To ensure the accuracy and diversity of book recommendations, a hybrid approach that integrates various technologies, including user interfaces, machine learning models, and recommendation algorithms, was employed.

The performance of the developed system was rigorously evaluated using key metrics such as accuracy, precision, recall, F1-score, and AUC. Impressively, the system exhibited outstanding accuracy (0.97) and achieved high precision, recall, and F1-score scores, demonstrating its effectiveness in providing relevant recommendations. Furthermore, the AUC score validated the system's capability to effectively differentiate between positive and negative recommendations.

In conclusion, the book exchange system with a hybrid recommendation engine proves its potential to offer accurate and suitable book recommendations to users, aligning with their unique interests and preferences. This solution not only addresses the limitations of traditional bookstores but also promotes sustainability, affordability, and a more engaging reading experience."