AI-Based Movie Recommendation System Using Collaborative Filtering

Abstract

AI-powered movie recommendation system integrates real-time sentiment analysis with sophisticated collaborative filtering techniques. The system leverages state-of-the-art natural language processing models for analyzing user mood inputs and combines this emotional context with comprehensive user preference tracking to deliver highly personalized movie suggestions. Built on a robust Flask-based architecture with MongoDB backend, the implementation demonstrates exceptional performance in understanding user emotions and generating contextaware recommendations. Experimental results show the system achieves 87.6% accuracy in sentiment classification and provides 42% more relevant recommendations compared to traditional approaches. The integration of real-time mood analysis and dynamic user profiling addresses critical limitations in existing recommendation systems, particularly in handling cold-start scenarios and adapting to evolving user preferences.