Abstract
Graph Neural Network (GNN) models have become increasingly popular for network analysis, especially in node classification tasks. However, the effectiveness of GNNs is compromised by two limitations. First, they implicitly assume that networks are homophilous, leading to decreased performance on heterophilous or random networks commonly found in the real world. Second, they tend to ignore the known node labels, inferring node labels merely from the node features and network structure. This is mainly rooted in the non-uniformity of node degrees, which makes it hard to directly aggregate label information. Hence, we propose a novel framework NRGNN, short for Graph Neural Network with Neighborhood Reconnection. NRGNN adjusts the network structure to increase homophily and uniformize node degrees. Then it applies message-passing-based or PageRank-based GNNs to the reconstructed networks, addressing the two limitations of GNNs. We evaluate NRGNN against 14 state-of-the-art baselines for node classification tasks. The empirical results demonstrate that NRGNN outperforms almost all the baselines, regardless of whether the datasets are homophilous, heterophilous, or random. The techniques might be adapted to more network analysis tasks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abu-El-Haija, S., et al.: MixHop: higher-order graph convolutional architectures via sparsified neighborhood mixing. In: Proceedings of the 36th International Conference on Machine Learning, pp. 21–29 (2019)
Al Shalabi, L., Shaaban, Z.: Normalization as a preprocessing engine for data mining and the approach of preference matrix. In: 2006 International Conference on Dependability of Computer Systems, pp. 207–214 (2006)
Chien, E., Peng, J., Li, P., Milenkovic, O.: Adaptive Universal Generalized PageRank Graph Neural Network (2021)
Gasteiger, J., Bojchevski, A., Günnemann, S.: Predict then Propagate: Graph Neural Networks meet Personalized PageRank (2022)
Hamilton, W., Ying, Z., Leskovec, J.: Inductive Representation Learning on Large Graphs. In: Advances in Neural Information Processing Systems. vol. 30 (2017)
Kipf, T.N., Welling, M.: Semi-Supervised Classification with Graph Convolutional Networks (2017)
Lim, D., Li, X., Hohne, F., Lim, S.N.: New Benchmarks for Learning on Non-Homophilous Graphs (2021)
Ma, Y., Liu, X., Shah, N., Tang, J.: Is homophily a necessity for graph neural networks? In: International Conference on Learning Representations (2022)
Maurya, S.K., Liu, X., Murata, T.: Improving Graph Neural Networks with Simple Architecture Design (2021)
Namata, G., London, B., Getoor, L., Huang, B.: Query-driven active surveying for collective classification. In: 10th International Workshop on Mining and Learning with Graphs. vol. 8, p. 1 (2012)
Pei, H., Wei, B., Chang, K.C.C., Lei, Y., Yang, B.: Geom-GCN: geometric graph convolutional networks. In: International Conference on Learning Representations (2019)
Qu, M., Bengio, Y., Tang, J.: GMNN: graph markov neural networks. In: Proceedings of the 36th International Conference on Machine Learning, pp. 5241–5250 (2019)
Rozemberczki, B., Allen, C., Sarkar, R.: Multi-scale attributed node embedding. J. Complex Netw. 9(2), cnab014 (2021)
Sen, P., Namata, G., Bilgic, M., Getoor, L., Galligher, B., Eliassi-Rad, T.: Collective classification in network data. AI Mag. 29(3), 93–93 (2008)
Şimşek, Ö., Jensen, D.: Navigating networks by using homophily and degree. Proc. Natl. Acad. Sci. 105(35), 12758–12762 (2008)
Tang, J., Sun, J., Wang, C., Yang, Z.: Social influence analysis in large-scale networks. In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 807–816. Paris France (2009)
Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., Bengio, Y.: Graph attention networks. In: International Conference on Learning Representations. arXiv:1710.10903 (2018)
Wang, H., Leskovec, J.: Unifying Graph Convolutional Neural Networks and Label Propagation (2020)
Wang, Y., Jin, J., Zhang, W., Yu, Y., Zhang, Z., Wipf, D.: Bag of Tricks for Node Classification with Graph Neural Networks (2021)
Xu, K., Li, C., Tian, Y., Sonobe, T., Kawarabayashi, K.I., Jegelka, S.: Representation learning on graphs with jumping knowledge networks. In: Proceedings of the 35th International Conference on Machine Learning, pp. 5453–5462 (2018)
Yan, Y., Hashemi, M., Swersky, K., Yang, Y., Koutra, D.: Two sides of the same coin: heterophily and oversmoothing in graph convolutional neural networks. In: 2022 IEEE International Conference on Data Mining (ICDM), pp. 1287–1292 (2022)
Yoshua, B., Olivier, D., Nicolas Le, R.: Label propagation and quadratic criterion. In: Chapelle, O., Scholkopf, B., Zien, A. (eds.) Semi-Supervised Learning, pp. 192–216. The MIT Press (2006)
Zhang, S., Liu, Y., Sun, Y., Shah, N.: Graph-less neural networks: teaching old MLPs new tricks via distillation. In: International Conference on Learning Representations (2022)
Zhang, X., Zitnik, M.: GNNGuard: defending graph neural networks against adversarial attacks. In: Advances in Neural Information Processing Systems, vol. 33, pp. 9263–9275 (2020)
Zhu, J., et al.: Graph neural networks with heterophily. Proc. AAAI Conf. Art. Intell. 35(12), 11168–11176 (2021)
Zhu, J., Yan, Y., Zhao, L., Heimann, M., Akoglu, L., Koutra, D.: Beyond homophily in graph neural networks: current limitations and effective designs. In: Advances in Neural Information Processing Systems. vol. 33, pp. 7793–7804 (2020)
Acknowledgement
Thanks to Sa Wang and Yungang Bao for their help. This work is supported in part by the Strategic Priority Research Program of the Chinese Academy of Sciences under grant numbers XDA0320000 and XDA0320300, the National Natural Science Foundation of China (Grant No. 62072433 and 62090022), the Fundamental Research Funds for the Central Universities (DUT21RC(3)102).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Guo, M., Sun, Z., Wang, Y., Liu, X. (2023). Graph Neural Network with Neighborhood Reconnection. In: Jin, Z., Jiang, Y., Buchmann, R.A., Bi, Y., Ghiran, AM., Ma, W. (eds) Knowledge Science, Engineering and Management. KSEM 2023. Lecture Notes in Computer Science(), vol 14117. Springer, Cham. https://doi.org/10.1007/978-3-031-40283-8_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-40283-8_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-40282-1
Online ISBN: 978-3-031-40283-8
eBook Packages: Computer ScienceComputer Science (R0)