Quantum Cybersecurity Analytics: Evaluation of Hybrid QML for DGA Botnet Detection
Open Access DepositedThis praxis presents a comprehensive investigation into the feasibility and speed of quantum cybersecurity analytics during the noisy intermediate scale quantum (NISQ) era. The research utilizes simulators and quantum devices from IonQ, Rigetti, Quantinuum, and the AER simulator. Classical and quantum binary classifiers are employed to explore the potential of quantum machine learning for achieving quantum advantage in cyber defense. Specifically, the praxis focuses on detecting domain generation algorithm (DGA) botnets, a technique commonly used in distributed denial-of-service attacks. Hybrid Quantum Binary Classifiers (HQBC) such as quantum neural networks, quantum support vector classifiers (QSVC), Pegasos: Primal estimated sub-gradient solver for SVC, and variational quantum classifiers (VQC) are tested on the DGA Botnet cybersecurity dataset. The research identifies and analyzes the respective limitations and bottlenecks of these classifiers.Furthermore, this research introduces a novel algorithm, the quantum Hoeffding tree classifier (QHTC). This algorithm applies the concept of online incremental learning to quantum algorithms through tree-based classification. Notably, QHTC outperforms existing algorithms in terms of both accuracy and speed, avoiding the optimization trap. The findings of this praxis suggest promising avenues for leveraging quantum computing capabilities in enhancing cybersecurity measures, even within the constraints of the NISQ device’s limitations.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
GolparvaranTehrani_gwu_0075A_16806.pdf | 2024-10-02 | Open Access |
|