Improved Graph Autoencoder for Network Reconstruction

Network embedding technology transforms network structure into node vectors, which reduces the complexity of representation and can be effectively applied to tasks such as classification, network reconstruction and link prediction. The most important concern of network embedding is how to retain the local structural features while effectively capturing the global features of the network. In view of the shortcomings of SDNE in weighted and directed network, this paper proposes an improved SDNE model based on node degree. The experimental results show that the improved SDNE model has better effect than the original algorithm with higher computational efficiency.