From Cluster Assumption to Graph Convolution: Graph-based Semi-Supervised Learning Revisited

Benchmark Model Rank Results
node-classification-on-citeseer-with-public-split-fixed-20-nodes-per-classOGC#1Accuracy: 77.5
node-classification-on-citeseer-with-public-split-fixed-20-nodes-per-classGGCM#10Accuracy: 74.2
node-classification-on-cora-with-public-split-fixed-20-nodes-per-classOGC#1Accuracy: 86.9%
node-classification-on-cora-with-public-split-fixed-20-nodes-per-classGGCM#20Accuracy: 83.6%
node-classification-on-pubmed-with-public-split-fixed-20-nodes-per-classOGC#1Accuracy: 83.4%
node-classification-on-pubmed-with-public-split-fixed-20-nodes-per-classGGCM#12Accuracy: 80.8%