Muhammad Abrar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Muhammad Abrar has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Mujeeb Ur Rehman
De Montfort University
3 shared publications - Sohail Khalid
University of Management and Technology
2 shared publications Imran Rehan
Gomal University
1 shared publication- Kamran Rehan
The University of Haripur
1 shared publication - Margaret Zita
De Montfort University
1 shared publication - Muhammad Kazim
School of Computer Science and Informatics, De Montfort University, Leicester LE1 9BH, UK
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 12182-12194
Quantum Machine Learning for Cybersecurity: Toward Smarter Intrusion Detection Systems
The growing scale and complexity of contemporary cyber threats require intrusion detection systems (IDS) that can operate effectively in high-dimensional feature spaces while maintaining robustness against rare and evolving attack patterns. Classical machine learning techniques, such as support vector machine (SVM) can face increasing limitations in modeling nonlinear decision boundaries under such conditions. This work presents a systematic and reproducible benchmark comparing classical support vector machine (SVM) and quantum support vector machine (QSVM) for nine-class network intrusion detection, covering benign traffic and eight attack categories (DoS Hulk, DDoS, FTP Brute Force, SSH Brute Force, SQL Injection, XSS, PortScan, and Infiltration). The evaluation is conducted on the CICIDS2017 dataset, comprising over 2.8 million labeled network flows. To ensure compatibility with near-term quantum hardware, mutual information (MI) based feature selection is employed to construct a compact yet information-preserving representation, which is encoded into a quantum feature map with an optimized entanglement structure. Experimental evaluation is conducted across two complementary settings. In the binary setting (1,000 balanced samples), classical SVM achieves 99.2% accuracy and 99.1% F1-score against QSVM’s 96.9% accuracy and 96.7% F1-score, an accuracy gap of 2.3 percentage points and an F1-score gap of 2.4 percentage points, reflecting marginal differences in precision–recall balance across the balanced test set. In the primary nine-class multiclass setting (1,800 balanced samples, 200 per class), classical SVM achieves a macro-F1 of 97.3% compared to QSVM’s 95.5%, narrowing the F1 gap to 1.8 percentage points. Critically, per-class analysis reveals that the SVM–QSVM gap narrows to only 0.3% for SQL Injection and Infiltration, the highest-complexity minority classes, suggesting that quantum kernel geometry provides relatively stronger representational capacity precisely for the rare, complex attack patterns of greatest operational significance. The proposed Docker- and Qiskit-based benchmarking framework provides a reproducible and extensible foundation for systematic classical–quantum benchmarking in cybersecurity, establishing a transparent baseline that can guide the design of future hybrid quantum–classical intrusion detection architectures.
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 12182-12194
Quantum Machine Learning for Cybersecurity: Toward Smarter Intrusion Detection Systems
The growing scale and complexity of contemporary cyber threats require intrusion detection systems (IDS) that can operate effectively in high-dimensional feature spaces while maintaining robustness against rare and evolving attack patterns. Classical machine learning techniques, such as support vector machine (SVM) can face increasing limitations in modeling nonlinear decision boundaries under such conditions. This work presents a systematic and reproducible benchmark comparing classical support vector machine (SVM) and quantum support vector machine (QSVM) for nine-class network intrusion detection, covering benign traffic and eight attack categories (DoS Hulk, DDoS, FTP Brute Force, SSH Brute Force, SQL Injection, XSS, PortScan, and Infiltration). The evaluation is conducted on the CICIDS2017 dataset, comprising over 2.8 million labeled network flows. To ensure compatibility with near-term quantum hardware, mutual information (MI) based feature selection is employed to construct a compact yet information-preserving representation, which is encoded into a quantum feature map with an optimized entanglement structure. Experimental evaluation is conducted across two complementary settings. In the binary setting (1,000 balanced samples), classical SVM achieves 99.2% accuracy and 99.1% F1-score against QSVM’s 96.9% accuracy and 96.7% F1-score, an accuracy gap of 2.3 percentage points and an F1-score gap of 2.4 percentage points, reflecting marginal differences in precision–recall balance across the balanced test set. In the primary nine-class multiclass setting (1,800 balanced samples, 200 per class), classical SVM achieves a macro-F1 of 97.3% compared to QSVM’s 95.5%, narrowing the F1 gap to 1.8 percentage points. Critically, per-class analysis reveals that the SVM–QSVM gap narrows to only 0.3% for SQL Injection and Infiltration, the highest-complexity minority classes, suggesting that quantum kernel geometry provides relatively stronger representational capacity precisely for the rare, complex attack patterns of greatest operational significance. The proposed Docker- and Qiskit-based benchmarking framework provides a reproducible and extensible foundation for systematic classical–quantum benchmarking in cybersecurity, establishing a transparent baseline that can guide the design of future hybrid quantum–classical intrusion detection architectures.
- Top Cited PapersOpen Access
- 7 January 2026
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 59 (2) , 40
The intersection of artificial intelligence and assistive technologies in the diagnosis and intervention of mental health conditions
Mental health disorders are becoming a major global health concern and pose a significant burden on global healthcare systems. Nearly one billion people suffer from mental disorders, accounting for 13% of the global disease burden and $1 trillion in annual productivity loss. Depression is the leading cause of disability and suicide is the second leading cause of death among young individuals. Economic uncertainty, social isolation, climate change, shifting societal norms, political conflict, and increasing violence are key factors contributing to the high prevalence of mental health issues. In the future, increasing poverty and inequality are likely to worsen this trend, resulting in a greater incidence and burden of mental illness. Therefore, timely diagnosis and intervention are a high priority. Traditional diagnostic and intervention methods, such as self-report questionnaires, clinical interviews, psychotherapy, medication, electroconvulsive therapy, and occupational therapy, have drawbacks including subjectivity, time commitment, and the potential for prolonged treatment. Due to these limitations, advanced approaches are needed to improve diagnostic accuracy and precision and to develop more effective interventions. This review aims to explore and evaluate the applications of Artificial Intelligence in the diagnosis and treatment of mental health conditions. This study provides a thorough analysis of various artificial intelligence-driven techniques and their advancements in the diagnosis of mental health conditions. Artificial intelligence has the potential to greatly improve the accuracy and effectiveness of mental health conditions. Moreover, this work consolidates the research gaps in current techniques and provides research hypotheses on how to overcome the gaps using a proposed 3-tier solution.
- 7 January 2026
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 59 (2) , 40
Top Cited PapersOpen AccessThe intersection of artificial intelligence and assistive technologies in the diagnosis and intervention of mental health conditions
Mental health disorders are becoming a major global health concern and pose a significant burden on global healthcare systems. Nearly one billion people suffer from mental disorders, accounting for 13% of the global disease burden and $1 trillion in annual productivity loss. Depression is the leading cause of disability and suicide is the second leading cause of death among young individuals. Economic uncertainty, social isolation, climate change, shifting societal norms, political conflict, and increasing violence are key factors contributing to the high prevalence of mental health issues. In the future, increasing poverty and inequality are likely to worsen this trend, resulting in a greater incidence and burden of mental illness. Therefore, timely diagnosis and intervention are a high priority. Traditional diagnostic and intervention methods, such as self-report questionnaires, clinical interviews, psychotherapy, medication, electroconvulsive therapy, and occupational therapy, have drawbacks including subjectivity, time commitment, and the potential for prolonged treatment. Due to these limitations, advanced approaches are needed to improve diagnostic accuracy and precision and to develop more effective interventions. This review aims to explore and evaluate the applications of Artificial Intelligence in the diagnosis and treatment of mental health conditions. This study provides a thorough analysis of various artificial intelligence-driven techniques and their advancements in the diagnosis of mental health conditions. Artificial intelligence has the potential to greatly improve the accuracy and effectiveness of mental health conditions. Moreover, this work consolidates the research gaps in current techniques and provides research hypotheses on how to overcome the gaps using a proposed 3-tier solution.
- 10 November 2025
- Published by Springer Nature
- p. 245-258
Zero-Day Attack Detection System Using Autoencoders and Isolation Forest: An Unsupervised Machine Learning Approach
- 10 November 2025
- Published by Springer Nature
- p. 245-258
Zero-Day Attack Detection System Using Autoencoders and Isolation Forest: An Unsupervised Machine Learning Approach