AI for Cybersecurity: Robust models for Authentication, Threat and Anomaly Detection DOI
Francesco Bergadano, Giorgio Giacinto, Olga Tushkanova

et al.

MDPI eBooks, Journal Year: 2023, Volume and Issue: unknown

Published: July 21, 2023

compare different ensemble learning methods that have been proposed in this context: Random Forests, XGBoost, CatBoost, GBM, and LightGBM.Experiments were performed on datasets, finding tree-based algorithms can achieve good performance with limited variability. Access Control [7,8]As stated above, access control be viewed as another point the anomaly detection continuum.Again, distinguishing a legitimate user from impostors automated through machine learning.The seventh paper [7] addresses context of face recognition systems (FRSs) proposes practical white box adversarial attack algorithm.The method is evaluated CASIA WebFace LFW datasets.In [8], authors used user's iris image, combined secret key, to generate public key subsequently use such data limit protected resources. Threat Intelligence [9,10]Not only do we want recognize block attacks they occur-we also need observe external overall network predict relevant events new patterns, addressing so-called threat intelligence landscape.In [9], two well-known databases (CVE MITRE) technique link correlate these sources.The tenth [10] formal ontologies monitor threats identify corresponding risks an way. ConclusionsIn conclusion, observed AI increasingly being cybersecurity, three main directions current research: (1) areas cybersecurity are addressed, CPS security intelligence; (2) more stable consistent results presented, sometimes surprising accuracy effectiveness; (3) presence AI-aware adversary recognized analyzed, producing robust reliable solutions.

Language: Английский

Cyberattacks in Smart Grids: Challenges and Solving the Multi-Criteria Decision-Making for Cybersecurity Options, Including Ones That Incorporate Artificial Intelligence, Using an Analytical Hierarchy Process DOI Creative Commons
Ayat-Allah Bouramdane

Journal of Cybersecurity and Privacy, Journal Year: 2023, Volume and Issue: 3(4), P. 662 - 705

Published: Sept. 27, 2023

Smart grids have emerged as a transformative technology in the power sector, enabling efficient energy management. However, increased reliance on digital technologies also exposes smart to various cybersecurity threats and attacks. This article provides comprehensive exploration of cyberattacks grids, focusing critical components applications. It examines cyberattack types their implications backed by real-world case studies quantitative models. To select optimal options, study proposes multi-criteria decision-making (MCDM) approach using analytical hierarchy process (AHP). Additionally, integration artificial intelligence (AI) techniques smart-grid security is examined, highlighting potential benefits challenges. Overall, findings suggest that “security effectiveness” holds highest importance, followed “cost-effectiveness”, “scalability”, “Integration compatibility”, while other criteria (i.e., “performance impact”, “manageability usability”, “compliance regulatory requirements”, “resilience redundancy”, “vendor support collaboration”, “future readiness”) contribute evaluation but relatively lower weights. Alternatives such “access control authentication” information event management” with high weighted sums are crucial for enhancing alternatives requirements” “encryption” still provide value respective criteria. We find “deep learning” emerges most effective AI technique “hybrid approaches”, “Bayesian networks”, “swarm intelligence”, “machine learning”, “fuzzy logic”, “natural language processing”, “expert systems”, “genetic algorithms” exhibit effectiveness addressing cybersecurity. The discusses drawbacks MCDM-AHP, enhancements its use cybersecurity, suggests exploring alternative MCDM evaluating options grids. aids decision-makers field make informed choices optimize resource allocation.

Language: Английский

Citations

40
Smart energy grid enabled by IoT: Potential uses and difficulties DOI

Shilpa Kalambe,

Bhavna Jain,

Pradhnya Morey

et al.

AIP conference proceedings, Journal Year: 2025, Volume and Issue: 3253, P. 020006 - 020006

Published: Jan. 1, 2025

Language: Английский

Citations

0
Enhancing cybersecurity via attribute reduction with deep learning model for false data injection attack recognition DOI Creative Commons
Faheed A. F. Alrslani, Manal Abdullah Alohali,

Mohammed Aljebreen

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 31, 2025

Language: Английский

Citations

0
Enhanced security data management in power grids using deep learning with genetic algorithm-based hyperparameter optimization DOI

Xianqi Cao

Peer-to-Peer Networking and Applications, Journal Year: 2025, Volume and Issue: 18(3)

Published: March 7, 2025

Language: Английский

Citations

0
Mitigating Missing Rate and Early Cyberattack Discrimination Using Optimal Statistical Approach with Machine Learning Techniques in a Smart Grid DOI Creative Commons

M. Nakkeeran,

V. Anantha Narayanan,

P. Bagavathi Sivakumar

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(8), P. 1965 - 1965

Published: April 20, 2024

In the Industry 4.0 era of smart grids, real-world problem blackouts and cascading failures due to cyberattacks is a significant concern highly challenging because existing Intrusion Detection System (IDS) falls behind in handling missing rates, response times, detection accuracy. Addressing this with an early attack mechanism reduced rate decreased time critical. The development Intelligent IDS vital mission-critical infrastructure grid prevent physical sabotage processing downtime. This paper aims develop robust Anomaly-based using statistical approach machine learning classifier discriminate from natural faults man-made events avoid failures. novel (SAML) based on Neighborhood Component Analysis, ExtraTrees, AdaBoost for feature extraction, bagging, boosting, respectively, proposed optimal hyperparameter tuning discrimination events. model tested publicly available Industrial Control Systems Cyber Attack Power (Triple Class) dataset three-bus/two-line transmission system Mississippi State University Oak Ridge National Laboratory. Furthermore, evaluated scalability generalization accessible IEEE 14-bus 57-bus datasets False Data Injection (FDI) attacks. test results achieved higher accuracy, lower false alarm compared approaches.

Language: Английский

Citations

2
Advanced mathematical modeling of mitigating security threats in smart grids through deep ensemble model DOI Creative Commons
Sanaa Sharaf, Mahmoud Ragab,

Nasser Albogami

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 4, 2024

Language: Английский

Citations

2
Enhancing DoS Detection in SmartGrid: Leveraging ML Using PCA and Explainable AI DOI

Sadman Saif,

Anika Tahsin Sarker,

Asad Islam

et al.

Published: May 2, 2024

Grid systems established integrated networks in every hierarchical level of power system for extensive smart automation. To ensure the availability and reliability SmartGrid system, fortification against cyber threats is crucial. Amidst other anomalies, Denial Service (DoS) disrupts normal network operations by overwhelming with excessive unauthorized traffic. assure cybersecurity undisrupted services, this article focuses on IEC 60 870-5-104 protocol concerning 12 classes DoS attack command detection. This anomaly detection mechanism utilizes a scrupulously indexed 60870-5-104 intrusion dataset through panoramic preprocessing subsequently introduces feature engineering dimension reduction Principal Component Analysis (PCA). Succeeding training ML models achieved raised definitive accuracy 98.709%. The introduction SHAP analysis presents unambiguous informative insight into model's decision-making most significant features. study establishes groundwork crafting robust security protocols that integrity operational stability face constantly evolving threats.

Language: Английский

Citations

1
Predicting Work-in-Process in Semiconductor Packaging Using Neural Networks: Technical Evaluation and Future Applications DOI Open Access

Chin-Ta Wu,

Shing‐Han Li,

David C. Yen

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(21), P. 4275 - 4275

Published: Oct. 31, 2024

This review paper focuses on the application of neural networks in semiconductor packaging, particularly examining how Back Propagation Neural Network (BPNN) model predicts work-in-process (WIP) arrival rates at various stages packaging processes. Our study demonstrates that BPNN models effectively forecast WIP quantities each processing step, aiding production planners optimizing machine allocation and thus reducing product manufacturing cycles. further explores potential applications enhancing efficiency, forecasting capabilities, process optimization within industry. We discuss integration real-time data from systems with network to enable more accurate dynamic planning. Looking ahead, this outlines prospective advancements for emphasizing their role addressing challenges rapidly changing market demands technological innovations. not only underscores practical implementations but also highlights future directions leveraging these technologies maintain competitiveness fast-evolving

Language: Английский

Citations

0
Swarming for success: Leveraging particle swarm optimization to enhance decision support in data mining DOI

Wasnaa Kadhim Jawad,

Abbas M. Al-Bakry

AIP conference proceedings, Journal Year: 2024, Volume and Issue: 3229, P. 040006 - 040006

Published: Jan. 1, 2024

Language: Английский

Citations

0
Enhancing Intrusion Detection Systems with Dimensionality Reduction and Multi-Stacking Ensemble Techniques DOI Creative Commons
Ali Mohammed Alsaffar, Mostafa Nouri-Baygi, Hamed M. Zolbanin

et al.

Algorithms, Journal Year: 2024, Volume and Issue: 17(12), P. 550 - 550

Published: Dec. 3, 2024

The deployment of intrusion detection systems (IDSs) is essential for protecting network resources and infrastructure against malicious threats. Despite the wide use various machine learning methods in IDSs, such often struggle to achieve optimal performance. key challenges include curse dimensionality, which significantly impacts IDS efficacy, limited effectiveness singular classifiers handling complex, imbalanced, multi-categorical traffic datasets. To overcome these limitations, this paper presents an innovative approach that integrates dimensionality reduction stacking ensemble techniques. We employ LogitBoost algorithm with XGBRegressor feature selection, complemented by a Residual Network (ResNet) deep model extraction. Furthermore, we introduce multi-stacking (MSE), novel method, enhance attack prediction capabilities. evaluation on benchmark datasets as CICIDS2017 UNSW-NB15 demonstrates our surpasses current models across performance metrics.

Language: Английский

Citations

0