GHRS: Graph-based hybrid recommendation system with application to movie recommendation

Research about recommender systems emerges over the last decade and comprises valuable services to increase different companies' revenue. Several approaches exist in handling paper systems. While most existing rely either on a content-based approach or collaborative approach, there are hybrid that can improve recommendation accuracy using combination of both approaches. Even though many algorithms proposed such methods, it is still necessary for further improvement. In this paper, we propose system method graph-based model associated with similarity users' ratings, demographic location information. By utilizing advantages Autoencoder feature extraction, extract new features based all combined attributes. Using set clustering users, our (GHRS) has gained significant improvement, which dominates other methods' performance cold-start problem. The experimental results MovieLens dataset show algorithm outperforms accuracy.