Semisupervised Graph Neural Networks for Graph Classification
- 27 April 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 53 (10) , 6222-6235
Abstract
Graph classification aims to predict the label associated with a graph and is an important graph analytic task with widespread applications. Recently, graph neural networks (GNNs) have achieved state-of-the-art results on purely supervised graph classification by virtue of the powerful representation ability of neural networks. However, almost all of them ignore the fact that graph classification usually lacks reasonably sufficient labeled data in practical scenarios due to the inherent labeling difficulty caused by the high complexity of graph data. The existing semisupervised GNNs typically focus on the task of node classification and are incapable to deal with graph classification. To tackle the challenging but practically useful scenario, we propose a novel and general semisupervised GNN framework for graph classification, which takes full advantage of a slight amount of labeled graphs and abundant unlabeled graph data. In our framework, we train two GNNs as complementary views for collaboratively learning high-quality classifiers using both labeled and unlabeled graphs. To further exploit the view itself, we constantly select pseudo-labeled graph examples with high confidence from its own view for enlarging the labeled graph dataset and enhancing predictions on graphs. Furthermore, the proposed framework is investigated on two specific implementation regimes with a few labeled graphs and the extremely few labeled graphs, respectively. Extensive experimental results demonstrate the effectiveness of our proposed semisupervised GNN framework for graph classification on several benchmark datasets.Keywords
Funding Information
- National Key Research and Development Program of China (2021ZD0113303)
- National Natural Science Foundation of China (62036006, 62106131, 62106134, 61906111, 61802317)
- Fund Program for the Scientific Activities of Selected Returned Overseas Professionals in Shanxi Province (20220002)
- Basic Research Program of Shanxi Province (20210302124032)
- Australian Research Council (LP180100114, DP200102611)
This publication has 27 references indexed in Scilit:
- Inductive Semi-supervised Multi-Label Learning with Co-Training2017
- node2vec2016
- Text Categorization as a Graph Classification Problem2015
- Label Propagation through Linear NeighborhoodsIEEE Transactions on Knowledge and Data Engineering, 2007
- Comparison of descriptor spaces for chemical compound retrieval and classificationKnowledge and Information Systems, 2007
- Shortest-Path Kernels on Graphs2005
- Protein function prediction via graph kernelsBioinformatics, 2005
- Distinguishing Enzyme Structures from Non-enzymes Without AlignmentsJournal of Molecular Biology, 2003
- Combining labeled and unlabeled data with co-training1998
- Probability of error of some adaptive pattern-recognition machinesIEEE Transactions on Information Theory, 1965