- paper: Adversarial Learning on Heterogeneous Information Networks
- Code from author: HeGAN
-
Clone the Openhgnn-DGL
python main.py -m HeGAN -d yelp4HeGAN -t node_classification -g -1
If you do not have gpu, set -gpu -1.
dataset | paper's | author's | ours |
---|---|---|---|
yelp(%) | 85.24 | 80.01 | 79.35 |
- We process the yelp dataset given by yelp. It saved as dgl.heterograph and can be loaded by dgl.load_graphs
-
Yelp
Dataset LinkEntity #Entity User 1,286 Business 2,614 Service 2 Star level 9 Reservation 2 Category 3
-
HeGAN
-
HeGAN is a wrapper of Generator and Discriminator.
-
Generator
-
A Discriminator
D
eveluates the connectivity between the pair of nodesu
andv
w.r.t. a relationr
. -
Discriminator
-
A generator
G
samples fake node embeddings from a continuous distribution. The distribution is Gaussian distribution.
Hui Han [GAMMA LAB]
Submit an issue or email to [email protected]