Recommending Movies on User's Current Preferences via Deep Neural Network
Muhammad Ali Faisal, Abdul Hameed, Akmal Saeed Khattak · 2019
Traditional recommendation system (RS) offers remarkable results in recommending movies. RS ignore the idea that preferences of a user changes with respect to time. The user cold case scenario is a problem in which user does not have a profile in the system. To address the cold case scenarios, we proposed, developed, and evaluated the recommendation engine based on user current preferences with the use of deep neural networks. The movies are fed to BGRU on the fly and a recommendation of the list of movies was made for the user to watch next. Our model recommends the movies based on his/her current preferences. The experiments were carried out on Movie Lens Dataset. The model was evaluated and have shown significant improvements in results in comparison to conventional RS. The results are presented based on Recall@K metrics resulting in accurate and personalized movies recommendation to user.