Weighted Graph Convolution Network for Multi-view Semi-supervised Classification

Recently, research on deep graph neural networks has received increasing attention. In some situations, the data exhibits an inherent graph structure. However, some multi-view data do not have an inherent graph structure, and there are few deep methods for multi-view data without graph structure. Based on the Graph Convolutional Network (GCN) architecture, we propose a sample weighted fusion graph semi-supervised classification (WFGSC) method suitable for multi-view data. Our experimental results demonstrate that the WFGSC model performs well.