A Data-Driven Hybrid Movie Recommendation System for Personalized Suggestions

The front end of the Movie Recommendation System is developed using Streamlit, providing an interactive and user-friendly interface for users to search and explore movies efficiently. The interface is enhanced with custom CSS styling to deliver a visually appealing and responsive experience through modern design elements such as animated cards and dynamic layouts. The back end is implemented using Python and integrates machine learning techniques to generate accurate recommendations. The system employs TF-IDF vectorization and cosine similarity for content-based filtering, allowing it to analyze movie features such as genres, keywords, cast, and overview. Additionally, collaborative filtering is implemented using Truncated Singular Value Decomposition (SVD) to capture user preferences and latent patterns from rating data. A hybrid approach is used by combining both methods to improve recommendation accuracy and personalization. The system utilizes datasets processed with Pandas and NumPy for efficient data handling. Furthermore, it integrates the OMDb API to fetch real-time movie details such as posters, ratings, and plot descriptions, enhancing user engagement. Streamlit session state and caching mechanisms are used to optimize performance and maintain seamless user interaction. This solution provides an efficient and scalable approach for personalized movie recommendations. In the future, the system can be extended with advanced machine learning models, user authentication, and deployment as a full-scale web or mobile application to support real-world usage.