Abstract
Model-level Graph Neural Network (GNN) explanation methods have become essential for understanding the decision-making processes of GNN models on a global scale. Many existing model-level GNN explanation methods often fail to incorporate prior knowledge of the original dataset into the initial explanation state, potentially leading to suboptimal explanation results that diverge from the real distribution of the original data. Moreover, these explainers often treat the nodes and edges within the explanation as independent elements, ignoring the structural relationships between them. This is particularly problematic in graph-based explanation tasks that are highly sensitive to structural information, which may unconsciously make the explanations miss key patterns important for the GNNs’ prediction. In this paper, we introduce KnowGNN, a knowledge-aware and structure-sensitive model-level GNN explanation framework, to explain GNN models in a global view. KnowGNN starts with a seed graph that incorporates prior knowledge of the dataset, ensuring that the final explanations accurately reflect the real data distribution. Furthermore, we construct a structure-sensitive edge mask learning method to refine the explanation process, enhancing the explanations’ ability to capture key features. Finally, we employ a simulated annealing (SA)-based strategy to control the explanation errors efficiently and thus find better explanations. We conduct extensive experiments on four public benchmark datasets. The results show that our method outperforms state-of-the-art explanation approaches by focusing explanations more closely on the actual characteristics of the data.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Data Availability
The datasets are publicly available online through their reference sources in the manuscript.
Code Availability
The code generated or used during the study appears in the submitted article.
Notes
References
Xu B, Huang J, Hou L, Shen H, Gao J, Cheng X (2020) Label-consistency based graph neural networks for semi-supervised node classification. In: 43rd International ACM SIGIR conference on research and development in information retrieval, pp. 1897–1900. ACM. https://doi.org/10.1145/3397271.3401308
Lin J, Wan Y, Xu J, Qi X (2023) Long-tailed graph neural networks via graph structure learning for node classification. Appl Intell 53(17):20206–20222. https://doi.org/10.1007/s10489-023-04534-3
Errica F, Podda M, Bacciu D, Micheli A (2020) A fair comparison of graph neural networks for graph classification. In: 8th International conference on learning representations. https://doi.org/10.48550/arXiv.1912.09893
Filtjens B, Vanrumste B, Slaets P (2022) Skeleton-based action segmentation with multi-stage spatial-temporal graph convolutional neural networks. IEEE Transactions on Emerging Topics in Computing, pp 1–11 (2022) https://doi.org/10.1109/TETC.2022.3230912
Tan R, Gao L, Khan N, Guan L (2022) Interpretable artificial intelligence through locality guided neural networks. Neural Netwo 155:58–73. https://doi.org/10.1016/j.neunet.2022.08.009
Zhou Y, Zhou T, Zhou T, Fu H, Liu J, Shao L (2021) Contrast-attentive thoracic disease recognition with dual-weighting graph reasoning. IEEE Transactions on Medical Imaging 40(4):1196–1206. https://doi.org/10.1109/TMI.2021.3049498
Grattarola D, Livi L, Alippi C, Wennberg R, Valiante TA (2022) Seizure localisation with attention-based graph neural networks. Expert Syst Appl 203:117330. https://doi.org/10.1016/j.eswa.2022.117330
Xu F, Qiao C, Zhou H, Calhoun VD, Stephen JM, Wilson TW, Wang Y (2023) An explainable autoencoder with multi-paradigm fmri fusion for identifying differences in dynamic functional connectivity during brain development. Neural Netw 159:185–197. https://doi.org/10.1016/j.neunet.2022.12.007
Crotti Junior A, Orlandi F, Graux D, Hossari M, O’Sullivan D, Hartz C, Dirschl C (2020) Knowledge graph-based legal search over german court cases. In: European semantic web conference. Cham, pp 293–297 . https://doi.org/10.1007/978-3-030-62327-2_44
Pradhan R, Lahiri A, Galhotra S, Salimi B (2022) Explainable ai: Foundations, applications, opportunities for data management research. In: Proceedings of the 2022 international conference on management of data. SIGMOD ’22, pp 2452–2457. Association for Computing Machinery. https://doi.org/10.1145/3514221.3522564
Tjoa E, Guan C (2021) A survey on explainable artificial intelligence (xai): Toward medical xai. IEEE Trans Neural Netw Learn Syst 32(11):4793–4813. https://doi.org/10.1109/TNNLS.2020.3027314
Ying R, Bourgeois D, You J, Zitnik M, Leskovec J (2019) GNNExplainer: Generating explanations for graph neural networks. In: Advances in neural information processing systems. https://proceedings.neurips.cc/paper_files/paper/2019/file/d80b7040b773199015de6d3b4293c8ff-Paper.pdf
Vu MN, Thai MT (2020) Pgm-explainer: Probabilistic graphical model explanations for graph neural networks. In: Advances in neural information processing systems, pp. 12225–12235. https://proceedings.neurips.cc/paper_files/paper/2020/file/8fb134f258b1f7865a6ab2d935a897c9-Paper.pdf
Huang Q, Yamada M, Tian Y, Singh D, Yin D, Chang Y (2020) GraphLIME: local interpretable model explanations for graph neural networks. IEEE Transactions on Knowledge and Data Engineering, pp 1–6. https://doi.org/10.1109/TKDE.2022.3187455
Yuan H, Yu H, Wang J, Li K, Ji S (2021) On explainability of graph neural networks via subgraph explorations. In: International conference on machine learning (ICML), pp 12241–12252. https://doi.org/10.48550/arXiv.2102.05152
Luo D, Cheng W, Xu D, Yu W, Zong B, Chen H, Zhang X (2020) Parameterized explainer for graph neural network. In: Advances in neural information processing systems, pp 19620–19631. https://proceedings.neurips.cc/paper_files/paper/2020/file/e37b08dd3015330dcbb5d6663667b8b8-Paper.pdf
Wang X, Wu Y, Zhang A, He X, Chua T-S (2021) Towards multi-grained explainability for graph neural networks. In: Advances in neural information processing systems, pp 18446–18458. https://proceedings.neurips.cc/paper_files/paper/2021/file/99bcfcd754a98ce89cb86f73acc04645-Paper.pdf
Schlichtkrull M.S, De Cao N, Titov I (2021) Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking. In: 9th International conference on learning representations. https://doi.org/10.48550/arXiv.2010.00577
Yuan H, Tang J, Hu X, Ji S (2020) Xgnn: Towards model-level explanations of graph neural networks. In: ACM SIGKDD international conference on knowledge discovery and data mining, pp 430–438. ACM. https://doi.org/10.1145/3394486.3403085
Wang X, Shen H-W (2023) Gnninterpreter: A probabilistic generative model-level explanation for graph neural networks. In: 11-th International conference on learning representations. https://doi.org/10.48550/arXiv.2209.07924
Sayan S, Monidipa D, Sanghamitra B (2023) Graphex: A user-centric model-level explainer for graph neural networks. In: 11-th International conference on learning representations. https://openreview.net/forum?id=CuE1F1M0_yR
Shin Y-M, Kim S-W, Shin W-Y (2024) Page: Prototype-based model-level explanations for graph neural networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, pp 1–18. https://doi.org/10.1109/TPAMI.2024.3379251
Azzolin S, Longa A, Barbiero P, Liò P, Passerini A (2023) Global explainability of gnns via logic combination of learned concepts. In: 11-th International conference on learning representations. https://doi.org/10.48550/arXiv.2210.07147
Xuanyuan H, Barbiero P, Georgiev D, Magister LC, Liò P (2023) Global concept-based interpretability for graph neural networks via neuron analysis. Proc AAAI Conf Artif Intell 37(9):10675–10683. https://doi.org/10.1609/aaai.v37i9.26267
Huang Z, Kosan M, Medya S, Ranu S, Singh A (2023) Global counterfactual explainer for graph neural networks. In: Proceedings of the sixteenth ACM international conference on web search and data mining. WSDM ’23, pp 141–149. Association for Computing Machinery. https://doi.org/10.1145/3539597.3570376
Ji Y, Shi L, Liu Z, Wang G (2024) Stratified gnn explanations through sufficient expansion. Proc AAAI Conf Artif Intell 38(11):12839–12847. https://doi.org/10.1609/aaai.v38i11.29180
Debnath AK, Debnath G, Hansch C, Compadre RLL, Shusterman AJ (1991) Structure-activity relationship of mutagenic aromatic and heteroaromatic nitro compounds. Correlation with Molecular Orbital Energies and Hydrophobicity. Journal of Medicinal Chemistry 34(2):786–797. https://doi.org/10.1021/jm00106a046
Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: 5th International conference on learning representations. https://openreview.net/forum?id=SJU4ayYgl
Bruna J, Zaremba W, Szlam A, LeCun Y (2014) Spectral networks and deep locally connected networks on graphs. In: 2nd International conference on learning representations. https://doi.org/10.48550/arXiv.1312.6203
Shuman DI, Narang SK, Frossard P, Ortega A, Vandergheynst P (2013) The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Signal Process Mag 30(3):83–98. https://doi.org/10.1109/MSP.2012.2235192
Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering. In: Advances in neural information processing systems, pp 3844–3852. https://proceedings.neurips.cc/paper_files/paper/2016/file/04df4d434d481c5bb723be1b6df1ee65-Paper.pdf
Xu B, Shen H, Cao Q, Qiu Y, Cheng X (2019) Graph wavelet neural network. In: 7th International conference on learning representations. https://openreview.net/forum?id=H1ewdiR5tQ
Hamilton WL, Ying R, Leskovec J (2017) Inductive representation learning on large graphs. In: Advances in neural information processing systems, pp 1025–1035. https://proceedings.neurips.cc/paper_files/paper/2017/file/5dd9db5e033da9c6fb5ba83c7a7ebea9-Paper.pdf
Chen J, Ma T, Xiao C (2018) FastGCN: Fast learning with graph convolu-tional networks via importance sampling. In: 6th International conference on learning representations. https://openreview.net/forum?id=rytstxWAW
Lucic A, Ter Hoeve M.A, Tolomei G, De Rijke M, Silvestri F (2022) Cf-gnnexplainer: Counterfactual explanations for graph neural networks. In: Camps-Valls G, Ruiz F.J.R, Valera I. (eds.) Proceedings of The 25th international conference on artificial intelligence and statistics. proceedings of machine learning research, vol 151, pp 4499–4511 . https://proceedings.mlr.press/v151/lucic22a.html
Veyrin-Forrer L, Kamal A, Duffner S, Plantevit M, Robardet C (2022) What does my gnn really capture? on exploring internal gnn representations. In: International joint conference on artificial intelligence 2022. https://doi.org/10.24963/ijcai.2022/105
Feng Q, Liu N, Yang F, Tang R, Du M, Hu X (2022) Degree: Decomposition based explanation for graph neural networks. In: 10-th International conference on learning representations. https://doi.org/10.48550/arXiv.2305.12895
Perotti A, Bajardi P, Bonchi F, Panisson A (2022) Graphshap: Motif-based explanations for black-box graph classifiers. https://doi.org/10.48550/arXiv.2202.08815
Zhang S, Liu Y, Shah N, Sun Y (2022) Gstarx: Explaining graph neural networks with structure-aware cooperative games. In: Advances in neural information processing systems, vol 35, pp 19810–19823. https://proceedings.neurips.cc/paper_files/paper/2022/file/7d53575463291ea6b5a23cf6e571f59b-Paper-Conference.pdf
Chhablani C, Jain S, Channesh A, Kash IA, Medya S (2024) Game-theoretic counterfactual explanation for graph neural networks. https://doi.org/10.48550/arXiv.2402.06030
Wu F, Li S, Jin X, Jiang Y, Radev D, Niu Z, Li SZ (2023) Rethinking explaining graph neural networks via non-parametric subgraph matching. In: Proceedings of the 40th International Conference on Machine Learning, vol. 202, pp. 37511–37523. https://doi.org/10.48550/arXiv.2301.02780
Lu S, Mills KG, He J, Liu B, Niu D (2024) Goat: Explaining graph neural networks via graph output attribution. In: 12-th International conference on learning representations. https://doi.org/10.48550/arXiv.2401.14578
Kang H, Han G, Park H (2024) Unr-explainer: Counterfactual explanations for unsupervised node representation learning models. In: 12-th International conference on learning representations. https://openreview.net/forum?id=0j9ZDzMPqr
Louizos C, Welling M, Kingma DP (2018) Learning sparse neural networks through L0 regularization. In: 6th International conference on learning representations. https://doi.org/10.48550/arXiv.1712.01312
Kingma DP, Welling M (2014) Auto-encoding variational bayes. In: 2nd International conference on learning representations. https://doi.org/10.48550/arXiv.1312.6114
Lu S, Yu Y, Yang J, Li B, Niu D (2024) Stexplainer: Global explainability of gnns via frequent subtree mining. In: 12-th International conference on learning representations. https://openreview.net/forum?id=HgSfV6sGIn
Wang X, Shen HW (2024) Gnnboundary: Towards explaining graph neural networks through the lens of decision boundaries. In: 12-th International conference on learning representations. https://openreview.net/forum?id=WIzzXCVYiH
Kirkpatrick S, Gelatt CD, Vecchi MP (1983) Optimization by simulated annealing. Science 220(4598):671–680. https://doi.org/10.1126/science.220.4598.671
Maddison C, Mnih A, Teh Y (2017) The concrete distribution: A continuous relaxation of discrete random variables. In: 5th International conference on learning representations. https://doi.org/10.48550/arXiv.1611.00712
Fisher ML (2004) The lagrangian relaxation method for solving integer programming problems. Manag Sci 50(12 SUPPL.):1861–1874. https://doi.org/10.1287/mnsc.1040.0263
Rozemberczki B, Allen C, Sarkar R (2021) Multi-Scale attributed node embedding. J Complex Netw 9(2):014. https://doi.org/10.1093/comnet/cnab014
Li Y, Zhou J, Verma S, Chen F (2022) A survey of explainable graph neural networks: Taxonomy and evaluation metrics. https://doi.org/10.48550/arXiv.2207.12599
Yuan H, Yu H, Gui S, Ji S (2023) Explainability in Graph Neural Networks: A Taxonomic Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence To appear 45(5):5782–5799. https://doi.org/10.1109/TPAMI.2022.3204236
Nguyen A, Clune J, Bengio Y, Dosovitskiy A, Yosinski J (2017) Plug and play generative networks: conditional iterative generation of images in latent space. In: IEEE conference on computer vision and pattern recognition, pp 3510–3520. https://doi.org/10.1109/CVPR.2017.374
Haynes WM, Lide DR, Bruno TJ (2015) CRC Handbook of Chemistry and Physics (95th Ed.). CRC Press,. https://doi.org/10.1201/9781315380476
Acknowledgements
This work was partially supported by the National Natural Science Foundation of China under grant 62072450. We also thank Dr. Tanvir Ahmad to refine the language of our manuscript and enhance the clarity of the result analysis.
Funding
The authors have no additional relevant funding conflicts.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Yinglong Ma: Conceptualization, Methodology, Writing-Reviewing, Editing, and Validation. Xiaofeng Liu: Writing Original draft preparation, Conceptualization, Coding, and Validation. Chenqi Guo: Investigation, Reviewing, and Validation. Beihong Jin: Methodology, Reviewing, and Validation. Huili Liu: Reviewing and Validation.
Corresponding author
Ethics declarations
Competing of Interest
The authors have no relevant financial or non-financial interests to disclose.
Ethics Approval and Consent to Participate
This work described in this manuscript is original and has not been under consideration for publication elsewhere. All authors read and approved the final manuscript. The research in this manuscript does not involve human participants and animals.
Consent for Publication
Written informed consent for publication was obtained from all participants.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Ma, Y., Liu, X., Guo, C. et al. KnowGNN: a knowledge-aware and structure-sensitive model-level explainer for graph neural networks. Appl Intell 55, 126 (2025). https://doi.org/10.1007/s10489-024-06034-4
Accepted:
Published:
DOI: https://doi.org/10.1007/s10489-024-06034-4