The DOI matches its Crossref record; 49 of the 53 references it cites were found in a registry or an index.
Anomaly Detection for Insider Threats Using Unsupervised Ensembles
Abstract
Insider threat represents a major cybersecurity challenge to companies, organizations, and government agencies. Insider threat detection involves many challenges, including unbalanced data, limited ground truth, and possible user behavior changes. This research presents an unsupervised learning based anomaly detection approach for insider threat detection. We employ four unsupervised learning methods with different working principles, and explore various representations of data with temporal information. Furthermore, different computational intelligence schemes are explored to combine these models to create anomaly detection ensembles for improving the detection performance. Evaluation results show that the approach allows learning from unlabelled data under challenging conditions for insider threat detection. Insider threats are detected with high detection and low false positive rates. For example, 60% of malicious insiders are detected under 0.1% investigation budget, and all malicious insiders are detected at less than 5% investigation budget. Furthermore, we explore the ability of the proposed approach to generalize for detecting new anomalous behaviors in different datasets, i.e., robustness. Finally, results demonstrate that a voting-based ensemble of anomaly detection can be used to improve detection performance as well as the robustness. Comparisons with the state-of-the-art confirm the effectiveness of the proposed approach.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Several studies [45], [49], [55], [57] , [62] have demonstrated that deep learning, characterized by the use of neural networks with multiple layers, has remarkable capabilities in capturing intricate patterns within complex datasets.”
From A Review of Recent Advances, Challenges, and Opportunities in Malicious Insider Threat Detection Using Machine Learning Methods (2024) · cites this paper as [55]
In the citing paper
“Scholars such as Le and Zincir-Heywood [52] focus on ML-based anomaly detection, while Yuan et al. emphasize DL-focused behavioral modeling [53].”
From Application of Machine Learning and Deep Learning Techniques for Enhanced Insider Threat Detection in Cybersecurity: Bibliometric Review (2025) · cites this paper as [52]
In the citing paper
“However, as noted by Le and Zincir-Heywood [52], these models struggled with dynamic and high-dimensional insider datasets, especially when it came to capturing sequential behavior.”
From Application of Machine Learning and Deep Learning Techniques for Enhanced Insider Threat Detection in Cybersecurity: Bibliometric Review (2025) · cites this paper as [52]
In the citing paper
“Scalability issues arise from the computational demands of processing vast datasets with Transformers, limiting realtime deployment with FAR up to 8% [21,27].”
From A systematic review on insider threat detection using natural language processing (2025) · cites this paper as [27]
In the citing paper
“Lack of realism is a recurring critique, with synthetic logs failing to capture dynamic, real-time threats like those in social media or chat logs, as highlighted in research gaps [12,27].”
From A systematic review on insider threat detection using natural language processing (2025) · cites this paper as [27]
2 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“AEs are trained using Adam optimization [49] for 100 epoch each.”
MethodScikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 2×
“Implementations from Scikit-learn [50] and PyOD [51] is used”
Method(2011) · Cited by 6,993
LOF: identifying density-based local outliers
Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000) · ACM SIGMOD Record · doi:10.1145/335191.335388 · Cited by 5,207
A Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection
Buczak, A. L., & Guven, E. (2016) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2015.2494502 · Cited by 3,332
Isolation-Based Anomaly Detection
Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2012) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/2133360.2133363 · Cited by 2,063 · cited 3×
“2) Isolation Forest (IF): Based on the principle that anomaly examples are rare and significantly different in attribute-values from normal data points, IF [40] is designed as an ensemble of “isolation trees”, whereas the anomalies – being easier to isolate – are assumed to be closer to the roots of the trees than nor…”
BackgroundInfluentialNetwork Anomaly Detection: Methods, Systems and Tools
Bhuyan, M. H., Bhattacharyya, D. K., & Kalita, J. K. (2014) · IEEE Communications Surveys & Tutorials · doi:10.1109/surv.2013.052213.00046 · Cited by 1,348 · cited 5×
“This is the initial and important detection step in cybersecurity workflow, where early signs of user behavior changes (anomaly) is flagged for further investigation, potentially to detect both known and unknown (zero day) attacks/vulnerabilities [7].”
BackgroundMethodInfluentialStatistical Comparisons of Classifiers over Multiple Data Sets
DemšarJanez (2006) · Journal of Machine Learning Research · doi:10.5555/1248547.1248548 · Cited by 1,074
PyOD: A Python Toolbox for Scalable Outlier DetectionPreprint
Zhao, Y., Nasrullah, Z., & Li, Z. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.01588 · Cited by 909 · cited 2×
“Implementations from Scikit-learn [50] and PyOD [51] is used”
MethodCan machine learning be secure?
Barreno, M., Nelson, B., Sears, R., Joseph, A. D., & Tygar, J. D. (2006) · Proceedings of the 2006 ACM Symposium on Information, computer and communications security · doi:10.1145/1128817.1128824 · Cited by 836
Research works printed but not linked: 7
- aggarwal, 2016, Outlier Analysis
- gavai, 2015, Supervised and unsupervised methods to detect insider threat from enterprise social and online activity data, J Wireless Mobile Netw Ubiquitous Comput Dependable Appl, 6, 47
- campagna, 2021, Enterprise insider threats on the rise
- 2014, Combating Insider Threats
- kent, 2015, Cybersecurity data sources for dynamic network research, Dynamic Networks in Cybersecurity
- abadi, 2015, TensorFlow Large-Scale Machine Learning on Heterogeneous Systems
- 2021, Insider Threat Test Dataset
Cited by
Artificial intelligence for cybersecurity: Literature review and future research directions
Kaur, R., Gabrijelčič, D., & Klobučar, T. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101804 · Cited by 615
Alzaabi, F. R., & Mehmood, A. (2024) · IEEE Access · doi:10.1109/access.2024.3369906 · Cited by 204
“…demonstrating efficacy in detecting subtle deviations indicative of insider threats [48], [50], [51], [76], Several studies [45], [49], [55], [57], [62] have demonstrated that deep learning, characterized by the use of neural networks with multiple layers, has remarkable capabilities…”
BackgroundMethodAl-Shehari, T., & Alsowail, R. A. (2021) · Entropy · doi:10.3390/e23101258 · Cited by 193
“…Markov models (HMM) 0.83 Al ‐ Mhiqani et al. [73] Deep neural network (DNN) 0.95 Gamachchi et al. [74] Attributed graph clustering (AGC) 0.76 [76] Unsupervised ensembles (UE) 0.91 Sharma et al. [77] Long short ‐ term memory (LSTM) 0.95 Singh et al. [78] Multi fuzzy classifier (MFC) 0.89 Wang…”
BackgroundInfluentialIntegrating machine learning for sustaining cybersecurity in digital banks
Asmar, M., & Tuqan, A. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e37571 · Cited by 98
“Anomaly detection algorithms can detect possible malicious actions carried out by insiders by tracking numerous data sources, such as log files, system events, or user activity logs [86].”
BackgroundMachine learning in identity and access management systems: Survey and deep dive
Aboukadri, S., Ouaddah, A., & Mezrioui, A. (2024) · Computers & Security · doi:10.1016/j.cose.2024.103729 · Cited by 91
Metadata
- Authors
- Le, D. C., & Zincir-Heywood, N.
- Affiliations
- Dalhousie University (CA)
- Venue
- IEEE Transactions on Network and Service Management
- Published
- 2021-06
- Volume
- 18
- Issue
- 2
- Pages
- 1152-1164
- ISSN
- 1932-4537 (ISSN-L)
- 2373-7379
- OpenAlex
- W3153493802
- Semantic Scholar
- 9447570eb330bd67110601aa2d8490c434e2d330
- Cited by
- 147
- References
- 53
- Citations by year
- 2026: 34 · 2025: 43 · 2024: 33 · 2023: 20 · 2022: 13 · 2021: 4
- FWCI
- 7.85
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Network and Service Management (publishedVersion)
- Funders
- Natural Science and Engineering Research Council of Canada (NSERC) and Compute Canada
- Killam Trusts and the province of Nova Scotia
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
- Physical Sciences › Computer Science › Information Systems › Information and Cyber Security
- Keywords
- insider threat detection, anomaly detection, unsupervised learning, ensemble learning, ensemble voting, temporal data representation, class imbalance, false positive reduction, user behavior analysis
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Anomaly Detection for Insider Threats Using Unsupervised Ensembles. https://app.cochat.ai/papers/10.1109/tnsm.2021.3071928
- The paper
- Le, D. C., & Zincir-Heywood, N. (2021). Anomaly Detection for Insider Threats Using Unsupervised Ensembles. IEEE Transactions on Network and Service Management. https://doi.org/10.1109/tnsm.2021.3071928
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.