Deep Discriminative Session-Based Recommender System

Due to the sequential nature and time-ordered session data, much research in a session-based recommender system (SBRS) focuses on recurrent neural networks (RNNs), including GRU and LSTM. On the other hand, convolutional neural networks (CNNs) provide very effective solutions for modeling sequential data when sequence elements are associated with complex features. As a result, we discuss different deep discriminative models in SBRS in this chapter, such as variants of RNNs and CNNs.