Abstract

Recommender systems have become essential in modern entertainment platforms, aiding users in discovering movies and TV shows tailored to their preferences. Traditional movie recommendation systems often rely on collaborative filtering or content-based methods. However, supervised learning offers an alternative approach by learning from labeled data, leveraging classification or regression models to predict user preferences. In this paper, we explore the implementation of a movie recommender system using supervised learning techniques, such as decision trees, random forests, and support vector machines. We discuss the dataset used, feature extraction techniques, model training, and evaluation metrics. The proposed system predicts user ratings and recommends movies based on historical data, user features, and movie characteristics. We demonstrate how supervised learning models can outperform traditional methods by integrating both user and item data in a structured manner.