Movie Recommendation System Using SVD (Letterboxd)

Abstract: The objective of this project is to create a movie recommendation system that utilizes Singular Value Decomposition (SVD) as the core algorithm. SVD is a well-known matrix factorization technique that effectively models user-movie interactions and extracts underlying features from the user-item matrix. By applying SVD, the recommendation system can uncover hidden patterns and similarities in movie preferences, resulting in precise and personalized movie recommendations. The system will utilize user ratings and movie metadata to construct a comprehensive user-item matrix, which will then undergo decomposition using SVD. The resulting low-rank approximation will be used to predict missing ratings and generate top-N movie recommendations for each user. The project will focus on optimizing the SVD algorithm, addressing data sparsity issues, and implementing an efficient recommendation generation process. The aim is to develop a scalable and accurate movie recommendation system that enhances user satisfaction, engagement, and overall movie-watching experiences.

Keywords: Single Value Decomposition, Movie Recommendation Works Cited: Aditya Bhardwaj, Chirla Rushil Reddy, Palak Arora " Movie Recommendation System Using SVD (Letterboxd) ", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 10, pp. 91-99, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.121013