The paper "Multi-view Denoising Graph Auto-Encoders on Heterogeneous Information Networks for Cold-start Recommendation" accepted by 2021 SIGKDD.
We evaluate our proposed method on three public benchmark datasets, i.e., DBook, MovieLens and Yelp, which are provided by https://github.com/rootlu/MetaHIN, including the training/testing data split.
We implement the proposed MvDGAE based on the large-scale graph learning framework PlatoDeep. Our code has been released within Tencent and will be pubished as the PlatoDeep released.