Graph Neural Networks for Intrusion Detection: A Survey
Open accessThe DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.
Abstract
Cyberattacks represent an ever-growing threat that has become a real priority for most organizations. Attackers use sophisticated attack scenarios to deceive defense systems in order to access private data or cause harm. Machine Learning (ML) and Deep Learning (DL) have demonstrate impressive results for detecting cyberattacks due to their ability to learn generalizable patterns from flat data. However, flat data fail to capture the structural behavior of attacks, which is essential for effective detection. Contrarily, graph structures provide a more robust and abstract view of a system that is difficult for attackers to evade. Recently, Graph Neural Networks (GNNs) have become successful in learning useful representations from the semantic provided by graph-structured data. Intrusions have been detected for years using graphs such as network flow graphs or provenance graphs, and learning representations from these structures can help models understand the structural patterns of attacks, in addition to traditional features. In this survey, we focus on the applications of graph representation learning to the detection of network-based and host-based intrusions, with special attention to GNN methods. For both network and host levels, we present the graph data structures that can be leveraged and we comprehensively review the state-of-the-art papers along with the used datasets. Our analysis reveals that GNNs are particularly efficient in cybersecurity, since they can learn effective representations without requiring any external domain knowledge. We also evaluate the robustness of these techniques based on adversarial attacks. Finally, we discuss the strengths and weaknesses of GNN-based intrusion detection and identify future research directions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Efficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568
Distributed Representations of Words and Phrases and their Compositionality
Mikolov, T., Sutskever, I., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1310.4546 · Cited by 24,533
A fast learning algorithm for deep belief nets
Hinton, G. E., Osindero, S., & Teh, Y.-W. (2006) · Neural Computation · doi:10.1162/neco.2006.18.7.1527 · Cited by 16,336
Neural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213
Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., & Bengio, Y. (2017) · arXiv · doi:10.48550/arxiv.1710.10903 · Cited by 12,107
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.3555 · Cited by 11,008
node2vec: Scalable Feature Learning for Networks
Grover, A., & Leskovec, J. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939754 · Cited by 10,926 · cited 1×
“In this section, we focus on two major random walk-based embedding techniques used in intrusion detection: DeepWalk [23] and node2vec [24].”
MethodInductive Representation Learning on Large Graphs
Hamilton, W. L., Ying, R., & Leskovec, J. (2017) · arXiv · doi:10.48550/arxiv.1706.02216 · Cited by 10,639
DeepWalk: online learning of social representations
Perozzi, B., Al-Rfou, R., & Skiena, S. (2014) · Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/2623330.2623732 · Cited by 10,123 · cited 1×
“In this section, we focus on two major random walk-based embedding techniques used in intrusion detection: DeepWalk [23] and node2vec [24].”
MethodThe graph neural network model
Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., & Monfardini, G. (2009) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2008.2005605 · Cited by 9,582 · cited 1×
“TheGraphNeural Network (GNN)model [36], [37] is the origin of the first application of deep neural networks to graphstructured data.”
BackgroundModel-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
Finn, C., Abbeel, P., & Levine, S. (2017) · arXiv · doi:10.48550/arxiv.1703.03400 · Cited by 8,512 · cited 1×
“The MAGNN was initially introduced to be used for the diagnosis of Alzheimer’s disease, where it outperforms most graph-based methods like GCN or GAT on biomedical datasets [91] by using a meta-learning paradigm [92].”
MethodModeling Relational Data with Graph Convolutional Networks
Schlichtkrull, M., Kipf, T. N., Bloem, P., van den Berg, R., Titov, I., & Welling, M. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-93417-4_38 · Cited by 5,570
Toward Generating a New Intrusion Detection Dataset and Intrusion Traffic Characterization
Sharafaldin, I., Habibi Lashkari, A., & Ghorbani, A. A. (2018) · Proceedings of the 4th International Conference on Information Systems Security and Privacy · doi:10.5220/0006639801080116 · Cited by 5,318 · cited 2×
“Experiments were conducted on the 80 original features, concluding that 8 selected features are sufficient to achieve valuable results [97].”
Background(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1953039 · Cited by 4,051
Moustafa, N., & Slay, J. (2015) · 2015 Military Communications and Information Systems Conference (MilCIS) · doi:10.1109/milcis.2015.7348942 · Cited by 4,033
Heterogeneous Graph Attention Network
Wang, X., Ji, H., Shi, C., Wang, B., Ye, Y., Cui, P., & Yu, P. S. (2019) · The World Wide Web Conference · doi:10.1145/3308558.3313562 · Cited by 3,191
One Pixel Attack for Fooling Deep Neural Networks
Su, J., Vargas, D. V., & Sakurai, K. (2019) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2019.2890858 · Cited by 2,406
A new model for learning in graph domains
Gori, M., Monfardini, G., & Scarselli, F. (2006) · Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005. · doi:10.1109/ijcnn.2005.1555942 · Cited by 2,075 · cited 2×
“Although deep learning on graphs has only been recently democratized, the first GNN [36] dates back to 2005 and is originally inspired from Recurrent Neural Networks (RNNs).”
BackgroundKoroniotis, N., Moustafa, N., Sitnikova, E., & Turnbull, B. (2019) · Future Generation Computer Systems · doi:10.1016/j.future.2019.05.041 · Cited by 1,983
DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning
Du, M., Li, F., Zheng, G., & Srikumar, V. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134015 · Cited by 1,835
Machine Learning and Deep Learning Methods for Intrusion Detection Systems: A Survey
Liu, H., & Lang, B. (2019) · Applied Sciences · doi:10.3390/app9204396 · Cited by 1,133 · cited 3×
“To counter these issues, Machine Learning (ML) has seen great success in cybersecurity tasks like intrusion detection [1], [2], [3] and malware detection [4], [5].”
BackgroundMethodInteraction Networks for Learning about Objects, Relations and Physics
Battaglia, P. W., Pascanu, R., Lai, M., Rezende, D., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1612.00222 · Cited by 969 · cited 1×
“The message function M is generally a differentiable function such as a neural network taking as input the concatenation of node and edge embeddings [39].”
BackgroundAn empirical comparison of botnet detection methods
García, S., Grill, M., Stiborek, J., & Zunino, A. (2014) · Computers & Security · doi:10.1016/j.cose.2014.05.011 · Cited by 962 · cited 4×
“The authors built a synthetic dataset containing benign and attack examples based on the CTU-13 dataset and Mirai attack’s distribution.”
BackgroundInfluentialGNNExplainer: Generating Explanations for Graph Neural Networks
Ying, R., Bourgeois, D., You, J., Zitnik, M., & Leskovec, J. (2019) · arXiv · doi:10.48550/arxiv.1903.03894 · Cited by 850
UNICORN: Runtime Provenance-Based Detector for Advanced Persistent Threats
Han, X., Pasquier, T., Bates, A., Mickens, J., & Seltzer, M. (2020) · Proceedings 2020 Network and Distributed System Security Symposium · doi:10.14722/ndss.2020.24046 · Cited by 349 · cited 3×
“Such graph representations capture important relationships between system events, which facilitates the discovery of malicious events that are temporally distant [118].”
BackgroundMethodBridging the Gap: A Pragmatic Approach to Generating Insider Threat Data
Glasser, J., & Lindauer, B. (2013) · 2013 IEEE Security and Privacy Workshops · doi:10.1109/spw.2013.37 · Cited by 344 · cited 1×
“CERT [148]: The CERT Insider Threat Test Dataset is a synthetic insider threat dataset provided by Carnegie Mellon University [149].”
MethodZhou, X., Liang, W., Li, W., Yan, K., Shimizu, S., & Wang, K. I.-K. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3130434 · Cited by 325 · cited 1×
“In paper [155], a novel approach is introduced to generate Hierarchical Adversarial Attacks (HAA) in order to specifically target GNN-based IDSs in Internet of Things (IoT) networks.”
MethodLiu, F., Wen, Y., Zhang, D., Jiang, X., Xing, X., & Meng, D. (2019) · Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3319535.3363224 · Cited by 319 · cited 3×
“Despite the current success of recent GNNs, random walks are still used in recent graph-based intrusion detection works [19], [20], [21], [22].”
BackgroundMethodRandom Walks: A Review of Algorithms and Applications
Xia, F., Liu, J., Nie, H., Fu, Y., Wan, L., & Kong, X. (2020) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2019.2952908 · Cited by 295 · cited 1×
“Random walks have been successfully applied to a large range of domains such as recommender systems and computer vision [18].”
BackgroundChen, T., Wang, Z., Li, L., Li, D., Li, Z., Chang, X., Bian, P., Liang, G., Wang, Q., & Xie, T. (2025) · Schloss Dagstuhl – Leibniz-Zentrum für Informatik · doi:10.4230/lipics.ecoop.2025.6
Cited by
Ngo, T., Yin, J., Ge, Y.-F., & Wang, H. (2025) · Information · doi:10.3390/info16060499 · Cited by 37
Metadata
- Authors
- Bilot, T., Madhoun, N. E., Agha, K. A., & Zouaoui, A.
- Venue
- IEEE Access
- Published
- 2023
- Volume
- 11
- Pages
- 49114-49139
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4376480418
- Cited by
- 227
- References
- 31
- FWCI
- 25.49
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.