Ali Elsadai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ali Elsadai has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Luka Jovanovic
Singidunum University
3 shared publications - Miodrag Zivkovic
Faculty of Informatics and Computing, Singidunum University, Danijelova 32, Belgrade 11000, Serbia
3 shared publications - Mohamed Salb
Singidunum University
3 shared publications - Nebojsa Bacanin
Singidunum University, Belgrade, Serbia
3 shared publications - Marko Sarac
Faculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, Serbia
2 shared publications - Aleksandar Sandro Cvetković
Faculty of Computing and Informatics, Sinergija University, Raje Banicica bb, 76300 Bijeljina, Bosnia and Herzegovina
1 shared publication
Latest Publications
- Open Access
- 30 September 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (10) , 306
A Reality Gap Taxonomy for Physics-Informed Anomaly Detection in Industrial Control Systems: Cross-Regime Evaluation, Operating-Point and Temporal Distribution Shifts
Physics-informed anomaly detectors for industrial control systems are usually validated on a single synthetic or testbed dataset, which can yield optimistic estimates that do not transfer to deployment-like conditions. This validation-to-deployment discrepancy, the reality gap, is normally treated as one phenomenon and attributed to generic domain shift. We argue that it comprises at least three mechanistically distinct failure modes and propose a taxonomy that separates them by what a practitioner can observe and act on: Type 1, cross-regime evaluation; Type 2, operating-point shift; and Type 3, temporal distribution shift. A single physics-informed framework combining a Kalman-filter validation engine, BiLSTM, β-VAE, and graph attention network with residual sharing is evaluated across HAI 23.05, SWaT, and WADI over five random seeds. The three cases behave differently. Type 1 shows a substantial cross-regime decrease: AI-only falls from 0.774 in the prior synthetic evaluation to 0.493 on HAI, and Full EADE from 0.798 to 0.496. Under the retrospective fixed-FPR comparison on HAI, GNN+BiLSTM has the highest mean F1 at 0.498, although it is not separable from Full_EADE across five seeds. Type 2 collapses to FPR = 0.850 with F1 ≈ 0.237 across configurations; a strongly shifted sensor gives the clearest localized signature, but the failure is driven by aggregate distributional change and is recoverable by global threshold recalibration. Type 3 shows moderate FPR (0.244) and the largest sensitivity to residual occlusion (ΔF1_occl = +0.086), although a trained residual-free ablation shows no corresponding loss on any dataset. The resulting taxonomy links distinct failure signatures to different deployment responses.
- 30 September 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (10) , 306
Open AccessA Reality Gap Taxonomy for Physics-Informed Anomaly Detection in Industrial Control Systems: Cross-Regime Evaluation, Operating-Point and Temporal Distribution Shifts
Physics-informed anomaly detectors for industrial control systems are usually validated on a single synthetic or testbed dataset, which can yield optimistic estimates that do not transfer to deployment-like conditions. This validation-to-deployment discrepancy, the reality gap, is normally treated as one phenomenon and attributed to generic domain shift. We argue that it comprises at least three mechanistically distinct failure modes and propose a taxonomy that separates them by what a practitioner can observe and act on: Type 1, cross-regime evaluation; Type 2, operating-point shift; and Type 3, temporal distribution shift. A single physics-informed framework combining a Kalman-filter validation engine, BiLSTM, β-VAE, and graph attention network with residual sharing is evaluated across HAI 23.05, SWaT, and WADI over five random seeds. The three cases behave differently. Type 1 shows a substantial cross-regime decrease: AI-only falls from 0.774 in the prior synthetic evaluation to 0.493 on HAI, and Full EADE from 0.798 to 0.496. Under the retrospective fixed-FPR comparison on HAI, GNN+BiLSTM has the highest mean F1 at 0.498, although it is not separable from Full_EADE across five seeds. Type 2 collapses to FPR = 0.850 with F1 ≈ 0.237 across configurations; a strongly shifted sensor gives the clearest localized signature, but the failure is driven by aggregate distributional change and is recoverable by global threshold recalibration. Type 3 shows moderate FPR (0.244) and the largest sensitivity to residual occlusion (ΔF1_occl = +0.086), although a trained residual-free ablation shows no corresponding loss on any dataset. The resulting taxonomy links distinct failure signatures to different deployment responses.
- Open Access
- 1 December 2024
- Published by Scientific Scholar in Journal of King Saud University – Science
- Vol. 36 (11)
Cloud spot instance price forecasting multi-headed models tuned using modified PSO
- 1 December 2024
- Published by Scientific Scholar in Journal of King Saud University – Science
- Vol. 36 (11)
Open AccessCloud spot instance price forecasting multi-headed models tuned using modified PSO
- 28 November 2023
- Published by Springer Nature
- p. 667-682
Cloud Computing Load Forecasting by Using Bidirectional Long Short-Term Memory Neural Network
- 28 November 2023
- Published by Springer Nature
- p. 667-682
Cloud Computing Load Forecasting by Using Bidirectional Long Short-Term Memory Neural Network
- 14 October 2022
- Published by Springer Nature
- p. 753-768
Training Logistic Regression Model by Enhanced Moth Flame Optimizer for Spam Email Classification
- 14 October 2022
- Published by Springer Nature
- p. 753-768
Training Logistic Regression Model by Enhanced Moth Flame Optimizer for Spam Email Classification
- 12 July 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 81 (25) , 36715-36733
New approach for fingerprint recognition based on stylometric features with blockchain and cancellable biometric aspects
- 12 July 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 81 (25) , 36715-36733
New approach for fingerprint recognition based on stylometric features with blockchain and cancellable biometric aspects
- Open Access
- 1 January 2019
- Published by Singidunum University
- p. 351-359
The Evaluation of The Technical Quality of Web Sites њith Contents in Arabic and Serbian Languages
In this paper, authors present the result of an experimental study which covers the technical evaluation of 200 websites
- 1 January 2019
- Published by Singidunum University
- p. 351-359
Open AccessThe Evaluation of The Technical Quality of Web Sites њith Contents in Arabic and Serbian Languages
In this paper, authors present the result of an experimental study which covers the technical evaluation of 200 websites
- Open Access
- 1 January 2019
- Published by Singidunum University
- p. 155-161
The Overview of Intrusion Detection System Methods and Techniques
System and network security is still the main concern for researchers and system administrators
- 1 January 2019
- Published by Singidunum University
- p. 155-161
Open AccessThe Overview of Intrusion Detection System Methods and Techniques
System and network security is still the main concern for researchers and system administrators