Hybrid Time Aware Recommender System combining CNN and RNN

Abstract—Recommender systems are often found in current e-commerce platforms to assist users in discovering suitable items or services. Traditional recommender systems, usually ignore the temporal dynamics of user-item interactions, leading to unsatisfying recommendations. We introduced the Hybrid Time Aware Recommender System (HTARS), a sophisticated recommendation a model that uses both Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) architectures to deliver personalized recommendations in time-sensitive circumstances. The model considers both user-item interactions and temporal changes in user preferences. The CNN component of the model oversees learning spatial characteristics of user-item interactions, while the RNN component captures their temporal relationships. Time-aware recommender systems have emerged as an intriguing answer to this problem.