MOVIE RECOMMENDATION SYSTEM USING MACHINE LEARNING

Abstract

Growing demand for tailored movie recommendation systems results from growing use of digital media. This research introduces an innovative approach that enhances recommendation accuracy by integrating collaborative filtering and content-based filtering methods. Collaborative filtering predicts user preferences based on historical interactions, whereas content-based filtering analyzes movie attributes. By merging these techniques, our hybrid system generates more precise and diverse recommendations. Additionally, an adaptive algorithm dynamically adjusts the influence of each filtering technique based on user engagement and item diversity. Performance evaluations indicate that our hybrid model surpasses conventional recommendation techniques in both accuracy and user satisfaction. The system's efficiency and scalability are validated using real-world datasets. This study contributes to advancing movie recommendation technologies by addressing existing limitations and providing insights for future improvements.