Topological Graph Convolutional Network Based on Complex Network Characteristics

Graph convolutional neural networks have received a lot of attention in various tasks dealing with graph data by aggregating information from neighboring nodes and passing node information. Many recent studies looked at the impact topological features on classification altering aggregation based degree values or incorporating analysis into networks; however, itself has many characteristics complex networks. In most circumstances, graph’s reveal nodes’ similarity facilitate task. This paper proposes structure feature extraction method concept characteristics, which can obtain deeper use to that is more important for both spaces. Evidence experimental established obtained this be used as input GCN,and good results achieved task even without any external nodes. dataset connected topologies, exhibits very large increase accuracy macro F1-score when compared state-of-the-art baseline model after mixing features.