A Movie Recommender System Based on Machine Learning and Deep Learning

Abstract

Digital entertainment platforms provide access to massive collections of movies and multimedia content, making personalized content discovery increasingly important. This work presents a hybrid movie recommendation system combining collaborative filtering based on ALS matrix factorization with content-based filtering using semantic embeddings and cosine similarity. The proposed system integrates web and mobile applications connected to a backend infrastructure and an intelligent recommendation engine capable of generating dynamic personalized recommendations. In addition, periodic retraining mechanisms were incorporated to continuously adapt recommendation models according to newly collected user interactions. Experimental results demonstrate the effectiveness of the proposed hybrid approach in improving recommendation relevance and personalization quality.