'Discrete Deep Learning for Fast Content-Aware Recommendation' published on ACM WSDM 2018 by authors Yan Zhang et al. DDL is a content-aware discrete recommendation for cold-start ans sparse recommendation.

Note: this version is a little different from the paper in which we use deep belief network(DBN) to learn the representations of content (in this code we use SAE). You can test the DBN performance just change SAE to DBN.

To run the code, you can just start with main_sae.m