Data science for next-generation recommender systems

Data science has been the foundation of recommender systems for a long time. Over past few decades, various have developed using different data and machine learning methodologies techniques. However, no existing work systematically discusses significant relationships between systems. To bridge this gap, paper aims to investigate from perspective science. Firstly, we introduce types used recommendations corresponding models methods that effectively represent each type. Next, provide brief outline representative utilized in building Subsequently, share some preliminary thoughts on next-generation Finally, summarize special issue