Deep Learning-Driven Personalized E-Book Recommendation System for E-Commerce Platforms

This study proposes an advanced e-book recommendation system that leverages deep learning models for accurate and dynamic user preference prediction. The framework integrates Attention Mechanism-based Long Short-Term Memory (AM-LSTM) with collaborative and content-based filtering to analyze user behavior, preferences, and past interactions. Additionally, the Rat Optimizer Algorithm (ROA) is employed to fine-tune LSTM weights, significantly improving classification accuracy. The model further incorporates Natural Language Processing (NLP) techniques for semantic analysis of user reviews and e-book metadata. Experimental results demonstrate higher accuracy, improved relevance, and enhanced user satisfaction compared to conventional recommendation approaches. This research highlights the potential of deep learning in refining recommendation systems, paving the way for more intelligent and user-centric e-commerce applications.