DenseNet-based attentive plot-aware recommendation

摘要

Since the ratings matrix is always sparse, auxiliary information has been proved very important in recommender systems. In this paper, we propose a DenseNet-based attentive plot-aware recommendation (DAPR) model, which combines attention mechanism and densely connected convolutional networks (i.e., DenseNet) to fully mine the semantic information in the movie plot text. This method effectively fuses rating information and text information for ratings prediction. Extensive experiments on three popular datasets demonstrate that our model performs better than other state-of-the-art approaches in common recommendation tasks.