Item-Based Collaborative Filtering for Movie Recommendation with Poster API Integration

With the rapid growth of streaming platforms, users are often overwhelmed by the vast number of movies and shows available, making it difficult to decide what to watch. Recommender systems help ease this challenge by offering personalized suggestions that match individual preferences, im- proving both user satisfaction and engagement. In this study, we develop a movie recommendation system based on Item-Based Collaborative Filtering (IBCF). The system analyzes relationships between movies using user–item interaction data and generates a list of top-N personalized recommendations. To make the experience more intuitive, the system integrates the TMDb API to display movie posters alongside suggestions, allowing users to quickly connect with and interpret their recommendations. The model is implemented on the MovieLens dataset and evaluated using metrics such as Root Mean Squared Error (RMSE), Precision@K, and Recall@K. Results show that IBCF offers both accuracy and scalability, while poster integration makes the interface more engaging. At the same time, challenges like the cold-start problem remain, which future work aims to address through hybrid models and deep learning approaches.