Content Filtering‐Based Movie Recommendation System Using Deep Learning
Atul K. Srivastava, Abhi Raj Singh · 2024
Movie recommendation systems are becoming increasingly popular as people seek personalized movie suggestions. In this work, a movie recommendation system that has been using machine learning has been proposed. Movie recommendation systems are widely used in online streaming platforms to improve user experience. We aim to develop a system that can provide accurate and personalized movie recommendations to users based on their behavior and preferences. The main problem addressed in this work is to improve the accuracy of recommendations based on historical data and preferences of the users. We aim to solve the cold start problem and increase the coverage of recommendations. A deep learning model based on content filtering has been developed to build a hybrid recommendation system. The performance of the proposed system is evaluated on a 5000-movie database available on IMDB. The model combines content-based and collaborative filtering techniques to offer personalized movie recommendations for users. We have also incorporated deep learning-based techniques to advance the precision and coverage of recommendations. Our results show that the hybrid recommendation system using content-based filtering and deep learning techniques can provide more accurate and personalized recommendations compared to traditional collaborative filtering algorithms. The implementation and results of our proposed system provide insights into the potential of machine learning in developing better movie recommendation systems. The proposed system outperforms other similar recommendation models based on filtering techniques. Incorporating deep learning-based techniques improved the accuracy of recommendations for users with sparse data. The proposed model can be deployed in the distributed environment.