An AI-Powered Mobile Application for Reading Management with a Hybrid Recommendation System

This paper presents the design and implementation of BookMark, an AI-based mobile reading management application developed to enhance user engagement and provide personalized reading experiences. The system first employs Optical Character Recognition (OCR) to extract text from book covers captured by the user and integrates external APIs to retrieve accurate bibliographic data. It then applies a GPT-based model to generate concise summaries of book descriptions and to create adaptive reading challenges across multiple difficulty levels. Furthermore, a KoBERT-based recommendation engine analyzes user preferences and reading patterns to suggest relevant titles, ensuring a customized reading journey. The overall system is implemented using React Native for mobile development and Firebase for backend services. By integrating these AI-driven functions, BookMark demonstrates an effective approach to intelligent reading support, offering users not only efficient access to information but also motivation to sustain continuous reading habits.