2020/09/28 by Mohammad Esmaeili, Esmaeili, Mohammad, Aria Nosratinia +2
Computer Science · Physics and Astronomy · #Advanced Graph Neural Networks #Complex Network Analysis Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Recommender Systems and Techniques #Social and Information Networks (cs.SI)
paper · pdf · doi:10.48550/arxiv.2009.13734
openalex publication_date 2020/09/28 · openalex created_date 2022/10/04 · openalex updated_date 2026/07/28
The nodes of a graph existing in a cluster are more likely to connect to each\nother than with other nodes in the graph. Then revealing some information about\nsome nodes, the structure of the graph (graph edges) provides this opportunity\nto know more information about other nodes. From this perspective, this paper\nrevisits the node classification task in a semi-supervised scenario by graph\nconvolutional networks (GCNs). The goal is to benefit from the flow of\ninformation that circulates around the revealed node labels. The contribution\nof this paper is twofold. First, this paper provides a method for extracting\nside information from a graph realization. Then a new GCN architecture is\npresented that combines the output of traditional GCN and the extracted side\ninformation. Another contribution of this paper is relevant to non-graph\nobservations (independent side information) that exists beside a graph\nrealization in many applications. Indeed, the extracted side information can be\nreplaced by a sequence of side information that is independent of the graph\nstructure. For both cases, the experiments on synthetic and real-world datasets\ndemonstrate that the proposed model achieves a higher prediction accuracy in\ncomparison to the existing state-of-the-art methods for the node classification\ntask.\n