b3ba8f1bee1238a2f37603d90b58898d-Reviews.html

First of all, we would like to thank the reviewers for their comments and insights. We have proposed to cast audio-based content recommendation as a latent factor regression problem using usage data as ground truth for training. We believe this is a novel idea, and by formulating the problem in this fashion we are able to build upon a large pre-existing body of research on regression techniques, as well as latent factor models. As a result, we of course agree with the reviewers that our paper does not contain any purely algorithmic machine learning contributions. However, we consider the fact that this idea can be applied with any kind of regression technique or latent factor model to be an advantage of our approach.