TY - JOUR
T1 - E-GCN: graph convolution with estimated labels
AU - Qin, Jisheng
AU - Zeng, Xiaoqin
AU - Wu, Shengli
AU - Tang, E
N1 - Funding Information:
This work was supported by the National Natural Science Foundation of China (No. 61170089, No. 60971088).
Publisher Copyright:
© 2021, Springer Science+Business Media, LLC, part of Springer Nature.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2021/1/6
Y1 - 2021/1/6
N2 - Graph Convolutional Network (GCN) has been commonly applied for semi-supervised learning tasks. How-ever, the established GCN frequently only considers the given labels in the topology optimization, which may not deliver the best performance for semi-supervised learning tasks. In this paper, we propose a novel Graph Convolutional Network with Estimated labels (E-GCN) for semi-supervised learning. The core design of E-GCN is to learn a suitable network topology for semi-supervised learning by linking both estimated labels and given labels in a centralized network frame-work. The major enhancement is that both given labels and estimated labels are utilized for the topology optimization in E-GCN, which assists the graph convolution implementation for unknown labels evaluation. Experimental results demonstrate that E-GCN is significantly better than state-of-the-art (SOTA) baselines without estimated labels.
AB - Graph Convolutional Network (GCN) has been commonly applied for semi-supervised learning tasks. How-ever, the established GCN frequently only considers the given labels in the topology optimization, which may not deliver the best performance for semi-supervised learning tasks. In this paper, we propose a novel Graph Convolutional Network with Estimated labels (E-GCN) for semi-supervised learning. The core design of E-GCN is to learn a suitable network topology for semi-supervised learning by linking both estimated labels and given labels in a centralized network frame-work. The major enhancement is that both given labels and estimated labels are utilized for the topology optimization in E-GCN, which assists the graph convolution implementation for unknown labels evaluation. Experimental results demonstrate that E-GCN is significantly better than state-of-the-art (SOTA) baselines without estimated labels.
KW - Semi-supervised Learning, Graph Convolution Topology, Optimization Estimated Labels
KW - Learning graph convolution
KW - Semi-supervised
KW - Topology optimization
KW - Estimated labels
UR - https://pure.ulster.ac.uk/en/publications/e-gcn-graph-convolution-with-estimated-labels
UR - http://www.scopus.com/inward/record.url?scp=85099062940&partnerID=8YFLogxK
U2 - https://doi.org/10.1007/s10489-020-02093-5
DO - https://doi.org/10.1007/s10489-020-02093-5
M3 - Article
SP - 1
EP - 9
JO - Applied Intelligence
JF - Applied Intelligence
SN - 0924-669X
ER -