The DOI matches its Crossref record; 47 of the 47 references it cites were found in a registry or an index.
DTITD: An Intelligent Insider Threat Detection Framework Based on Digital Twin and Self-Attention Based Deep Learning Models
Open accessAbstract
Latest statistics and studies shows that the loss generated by insider threats is much higher than the external attacks. More and more organizations are increasing invest or purchase insider threat detection system to prevent insider risks. However, accurately and timely detecting insider threat face huge challenges. In this paper, we proposed an intelligent insider threat detection framework based on Digital Twin and self-attentions based deep learning model. First, this paper introduced what the insider threats are and the challenges of detecting them. Then this paper showed related recent works on solving insider threat detection problems and their limitations. Next, this paper proposed our solutions to address these challenges: building the innovative intelligent insider threat detection framework based on Digital Twin (DT) and self-attention based deep learning models, performing insight analysis of users’ behavior and entities, adopting contextual word embedding techniques using Bidirectional Encoder Representations from Transformers (BERT) model and sentence embedding technique using Generative Pre-trained Transformer 2 (GPT-2) model to make data augmentation to overcome significant data imbalance, and adopting temporal semantic representation of users’ behaviors to build user behavior time sequence. After that, this paper built self-attention based deep learning models to quickly detect insider threat. This paper proposed a simplified transformer model named DistilledTrans and applied original transformer model, DistilledTrans, BERT + final layer, Robustly Optimized BERT Approach (RoBERTa) + final layer, and the hybrid method combining pre-trained (BERT, RoBERTa) with Convolutional Neural Network (CNN) or Long Short-term Memory (LSTM) network model to detect insider threats. Finally, this paper showed experiment results on dense dataset CERT r4.2 and augmented sporadic dataset CERT r6.2, evaluated their performance and made comparison analysis with the state-of-the-art models. Promising experimental results shows that 1) contextual word embedding insert and substitution predicted by BERT model, and context embedding sentence predicted by GPT-2 model are effective data augmentation approaches to address highly data imbalance 2) DistilledTrans trained with sporadic dataset CERT r6.2 augmented by contextual embedding sentence method predicted by GPT-2 outperforms the state-of-the-art models in term of all evaluation metrics including accuracy, precision, recall, F1-score, and AUC. Additionally, its structure is much simpler and thus training time and computing cost are much less than the recent models 3) When trained with the dense dataset CERT r4.2, Pre-trained models BERT plus a final layer or RoBERTa plus a final layer can get significantly higher performance than the current models with a very little sacrifice of precision. In compassion, complex hybrid methods may not be necessary.
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
Figures







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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Challenges include high computational costs, as DL models require significant resources for training on large datasets like Los Alamos cybersecurity events [23,29], limiting real-time deployment [24,31].”
From A systematic review on insider threat detection using natural language processing (2025) · cites this paper as [23]
Experimental — accuracy not yet measured
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
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 1×
“The transformer model was introduced in 2017 by a team at Google Brain [30] and is increasingly becoming one of the first choices in Natural Language Processing (NLP).”
MethodExploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173 · cited 1×
“The BERT model [31] is a pretrained transformer-based model developed by Google.”
MethodBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 1×
“The BERT model [31] is a pretrained transformer-based model developed by Google.”
MethodRoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515 · cited 1×
“Robustly Optimized BERT Approach (RoBERTa) [32] was developed by Facebook AI.”
MethodDigital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems
Grieves, M., & Vickers, J. (2017) · Transdisciplinary Perspectives on Complex Systems · doi:10.1007/978-3-319-38756-7_4 · Cited by 3,727 · cited 1×
“The digital twin (DT) concept was introduced by Grieves et al. [28] in 2002 as an integrated probabilistic simulation of a physical asset or system to emulate the lifetime of its matching physical twin by using the best existing physical model, history information, latest sensor data updates, etc.…”
MethodDigital Twins: The Convergence of Multimedia Technologies
El Saddik, A. (2018) · IEEE MultiMedia · doi:10.1109/mmul.2018.023121167 · Cited by 817 · cited 1×
“…a physical asset or system to emulate the lifetime of its matching physical twin by using the best existing physical model, history information, latest sensor data updates, etc. DT was renewed by El Saddik and Abdulmotaleb [29] in 2018 as a digital replica of a living or non-living physical entity.”
BackgroundLiu, 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
RTIDS: A Robust Transformer-Based Approach for Intrusion Detection System
Wu, Z., Zhang, H., Wang, P., & Sun, Z. (2022) · IEEE Access · doi:10.1109/access.2022.3182333 · Cited by 317 · cited 1×
“In contrast to some scholar’s [33] conclusion, this study found that adding or reducing one encoder/decoder layer or putting a normalization layer before or after the multi-head self-attention mechanism or FFN achieved almost the same performance.”
ResultBeehive: large-scale log analysis for detecting suspicious activity in enterprise networks
Yen, T.-F., Oprea, A., Onarlioglu, K., Leetham, T., Robertson, W., Juels, A., & Kirda, E. (2013) · Proceedings of the 29th Annual Computer Security Applications Conference · doi:10.1145/2523649.2523670 · Cited by 270 · cited 1×
“Beehive [16] applied PCA to reduce the dimensions of features and then chose 15 features per day per host.”
Method
Research works printed but not linked: 6
- gamachchi, 2017, Graph based framework for licious insider threat detection, Proc 50th Hawaii Int Conf St Sci, 2638
- 2020, Market Guide for Insider Risk Management Solutions Jonathan Care Paul Furtado Brent Predovich USA
- 2022, Market Guide for Insider Risk Management Solutions Jonathan Care Paul Furtado Brent Predovich USA
- 2020, Insider Risk Report DTEX Systems USA
- 2018, 2018 U S state of Bercrime
- 2019, Insider Threat Report 2019
Cited by
Network Digital Twin for 6G and Beyond: An End-to-End View Across Multi-Domain Network Ecosystems
Tran, D.-H., Waheed, N., Saputra, Y. M., Lin, X., Nguyen, C. T., Abdu, T. S., Vo, V. N., Pham, V., Alsenwi, M., Adam, A. B. M., et al. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3599866 · Cited by 78
“Totackle insider threats, the authors in [126] introduce a novel framework for insider threat detection using DT and DL models with self-attentions.”
Alhumam, N., Rahman, M. M. H., & Aljughaiman, A. (2025) · IEEE Access · doi:10.1109/access.2025.3545004 · Cited by 65
Insider threat detection for specific threat scenarios
Tian, T., Zhang, C., Jiang, B., Feng, H., & Lu, Z. (2025) · Cybersecurity · doi:10.1186/s42400-024-00321-w · Cited by 42
Innovative AI strategies for enhancing smart building operations through digital twins: A survey
Oulefki, A., Kheddar, H., Amira, A., Kurugöllü, F., Himeur, Y., & Bounceur, A. (2025) · Energy and Buildings · doi:10.1016/j.enbuild.2025.115567 · Cited by 37
Wang, Z. Q., Wang, H., & Saddik, A. E. (2024) · IEEE Access · doi:10.1109/access.2024.3482988 · Cited by 41
“To demonstrate the superior generalization ability of Fed-ITD, we conducted data processing before the experiment such as DistilledTrans [7], LSTM-RNN [57], DeepMIT [58], DD-GCN [36], multistate LSTM + CNN [59], Hierarchical LSTMs [32], simultaneous neural learning [60], unsupervised ensembles…”
BackgroundMethodResultInfluentialSelf-citation
Metadata
- Authors
- Wang, Z. Q., & El Saddik, A.
- Affiliations
- University of Ottawa (CA)
- Multimedia Communications Research Laboratory (MCRLab), School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, Canada
- Venue
- IEEE Access
- Published
- 2023
- Volume
- 11
- Pages
- 114013-114030
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4387609283
- Semantic Scholar
- 0aaba06370eda5751c385c15a269728fba47ad4b
- Cited by
- 49
- References
- 47
- Citations by year
- 2026: 17 · 2025: 21 · 2024: 11
- FWCI
- 11.50
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- IEEE Access (publishedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Information Systems › Information and Cyber Security
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Keywords
- insider threat detection, digital twin, self-attention, BERT, GPT-2, data augmentation, class imbalance, 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: DTITD: An Intelligent Insider Threat Detection Framework Based on Digital Twin and Self-Attention Based Deep Learning Models. https://app.cochat.ai/papers/10.1109/access.2023.3324371
- The paper
- Wang, Z. Q., & El Saddik, A. (2023). DTITD: An Intelligent Insider Threat Detection Framework Based on Digital Twin and Self-Attention Based Deep Learning Models. IEEE Access. https://doi.org/10.1109/access.2023.3324371
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.