Music Recommendation with Temporal Dynamics in Multiple Types of User Feedback

In the music streaming service, users can feedback their preference explicitly or implicitly. We propose a link prediction approach for music recommendation with various types of user feedback to alleviate the data sparsity problem and the cold-start problem. Moreover, by reflecting the temporal features of the feedback, we analysis the time-varying user taste. The experiment on real-world dataset demonstrates the effectiveness of our approach in the recommendation quality.