基于卷积图神经网络的多粒度表示学习框架
张蕾, 钱峰, 赵姝, 陈洁, 杨雪洁, 张燕平

Multi⁃granular representation learning framework for Convolutional Graph Neural Networks
Lei Zhang, Feng Qian, Shu Zhao, Jie Chen, Xuejie Yang, Yanping Zhang
表3 半监督节点分类实验结果对比
Table 3 Experimental results of semi?supervised node classification by different algorithms
方法指标CoraCiteSeerDBLPPubMed
GCNMi⁃F10.8340.7360.8250.830
Ma⁃F10.8080.6830.7670.827

M⁃NRL

(GCN)

Mi⁃F10.8490.7430.8280.855
Ma⁃F10.8390.6880.7810.850

Graph

SAGE

Mi⁃F10.8240.7290.8150.841
Ma⁃F10.8000.6820.7590.837

M⁃NRL

(SAGE)

Mi⁃F10.8450.7460.8220.856
Ma⁃F10.8370.6890.7720.852
GATMi⁃F10.8370.7310.8300.848
Ma⁃F10.8160.6870.7840.842

M⁃NRL

(GAT)

Mi⁃F10.8400.7440.8370.857
Ma⁃F10.8280.6880.7930.852
APPNPMi⁃F10.8320.7420.8110.826
Ma⁃F10.8130.6740.7050.824

M⁃NRL

(APPNP)

Mi⁃F10.8400.7490.8280.861
Ma⁃F10.8320.6800.7780.858
TAGCNMi⁃F10.8540.7300.8280.843
Ma⁃F10.8420.6700.7740.839

M⁃NRL

(TAGCN)

Mi⁃F10.8630.7480.8320.856
Ma⁃F10.8470.6960.7770.853
GFNNMi⁃F10.8590.7320.8400.828
Ma⁃F10.8510.6700.7880.820

M⁃NRL

(GFNN)

Mi⁃F10.8680.7520.8500.855
Ma⁃F10.8530.6940.8230.851