About adding KGCN as an example of Recommender Systems

dmlc / dgl

Python package built to ease deep learning on graph, on top of existing DL frameworks.

http://dgl.ai

Apache License 2.0

13.55k stars 3.01k forks source link

About adding KGCN as an example of Recommender Systems #1138

Open pengfei opened 4 years ago

pengfei commented 4 years ago

🚀 Feature

Can kgcn be used as an example of a recommendation system? Examples of recommendation systems are few. KGCN involves neighbor sampling, which is a good example of Nodeflow. It's also not too complicated Hope to implement with Pytorch.

(www19)Knowledge Graph Convolutional Networks for Recommender Systems https://arxiv.org/abs/1904.12575 https://github.com/hwwang55/KGCN

zzhang-cn commented 4 years ago

I agree. Would you like to take a lead on it and we can assign someone to help?

pengfei commented 4 years ago

I am a beginner of DGL, and it may be more suitable for others who are more familiar with DGL to implement.

jermainewang commented 4 years ago

Currently, we are lacking the bandwidth on the whole DGL+Recsys direction. Labeled this with help wanted.