INTEGRATING SENTIMENT ANALYSIS AND MACHINE LEARNING FOR IMPROVED FILM RECOMMENDATION SYSTEMS

Personalization has emerged as a crucial element for boosting user satisfaction in the swiftly changing realm of digital entertainment platforms. This project introduces a sentiment-aware movie recommendation system designed to offer highly customized suggestions to movie enthusiasts, particularly targeting audiences such as Cinemaniacs and Film Junkies. The system merges collaborative filtering techniques with natural language processing (NLP) methods to conduct sentiment analysis of user-generated reviews and ratings. By integrating user preferences such as genre selection and viewing history with sentiment insights derived from textual movie reviews, the platform transcends traditional numeric ratings to capture more nuanced indicators of user satisfaction and dissatisfaction. The recommendation engine was enhanced with genre-specific filtering, actor-based matching, and predictive modelling to anticipate interest in upcoming movies and web series releases. Sentiment analysis is driven by a fine-tuned BERT model, which improves the accuracy of emotional tone detection in reviews. In addition, the system offers trailer predictions and releases alerts to keep users informed and engaged with emerging content. This multilayered architecture aims to deliver a more intelligent and emotionally responsive user experience, ultimately fostering deeper engagement by aligning recommendations with both behavioral and emotional user patterns.