Neural Collaborative Filtering‐Based Hybrid Recommender System for Online Movies Recommendation
S. Priyanka, Prathik Saravanan, Vairavasundaram Indragandhi, V. Subramaniyaswamy · 2023
A recommendation process is usually the most commonly used form of commercial website. The custom recommender method is of vital importance in modeling users’ choice of movies based on their previous interest. In today's internet apps, recommender services play a crucial role. However, the existing Deep Learning (DL) approaches have some limitations, which have a negative impact on the efficiency of the suggestion models. This paper provides an innovative deep learning architecture to improve filtering results in recommender systems. This methodology proposes new movies for users based on individual and related tastes. By studying user trends of film watching, we offer a technique of anticipating and suggesting a film. The similarity between any set of users is estimated using movie rating information and review data. Further we classified the user with similar film preferences and analysed the user group's consumption behavior to increase forecast accuracy by factoring the change in preferences over time. As films are an important source of entertainment, we have proposed a Recommender System (RS) in this work. Collaborative filtering and content-based filtering methods provide a conventional approach to recommendation systems with sentiment restrictions. We use a Hybrid filtering system based on simple Recurrent Neural Network (RNN) to demonstrate the efficiency of the suggested technique. The proposed system outperforms in terms of Root Mean Square Error (RMSE) and Mean Absoulte Error.