面向节点分类的图神经网络节点嵌入增强模型
|
|
曾菊香,王平辉,丁益东,兰林,蔡林熹,管晓宏
|
Graph neural network based node embedding enhancement model for node classification
|
|
Ju-xiang ZENG,Ping-hui WANG,Yi-dong DING,Lin LAN,Lin-xi CAI,Xiao-hong GUAN
|
|
| 表 4 Pubmed数据集上的分类准确度 |
| Tab.4 Prediction accuracy on Pubmed dataset % |
|
| 方法 | 5节点/类 | 10节点/类 | 20节点/类 | | Raw | 57.7±1.3 | 62.3±3.3 | 66.5±2.4 | | MLP | 59.8±4.1 | 64.3±3.2 | 69.6±2.6 | | DeepWalk | 54.5±4.7 | 60.7±4.2 | 65.1±3.5 | | ChebyNet | 65.4±7.0 | 70.1±4.9 | 75.0±2.5 | | GCN | 68.2±6.2 | 72.9±3.8 | 77.0±2.4 | | GAT | 71.3±5.3 | 74.7±2.9 | 77.7±2.5 | | GCN-DSM | *68.8±5.7 | *73.8±3.7 | 77.0±2.5 | | GAT-DSM | *71.8±5.2 | *75.6±3.1 | *78.7±2.2 |
|
|
|