Jason Hartley is lecturer in criminology at Griffith University in Brisbane, Australia. He is a former police officer with 23 years of experience, and has trained personnel for deployment in Timor Leste, the Solomon Islands, Iraq and Afghanistan. Jason specializes in, and has published on engagement with Muslim communities, Indigenous Polynesian approaches to rehabilitation and reducing recidivism, and Asian Organised Crime. Jason also completed a community internship in Hebron on the West Bank.
ITC522 Machine Learning for Cyber Threat Detection
This unit explores the application of machine learning techniques in the detection and mitigation of cyber threats. Students will learn how machine learning algorithms can be employed to identify patterns, anomalies, and malicious activities within large datasets, improving the security of networks, systems, and data. The unit covers supervised and unsupervised learning methods, neural networks, and deep learning, focusing on their use in cyber threat detection systems. Students will also examine real-world applications, including malware detection, intrusion detection systems, and network traffic analysis.
RELEVANT COURSES
* Core unit
CREDIT POINTS
10
STUDY MODES
On campus, online, hybrid
PREREQUISITE OR CO-REQUISITE
ITC404 Foundations of Cybersecurity
UNIT LEARNING OUTCOMES
- Critically review and interpret machine learning algorithms and their application in global cybersecurity contexts
- Critically assess machine learning models to detect cyber threats, applying metrics to measure effectiveness
- Critically evaluate and design machine learning models for specific cybersecurity threats, employing data preprocessing, feature engineering, and algorithm selection to enhance detection accuracy
- Design and implement machine learning-based cybersecurity solutions, integrating threat detection, anomaly identification, and performance evaluation to safeguard systems and data
- Articulate emerging global trends in machine learning for cybersecurity, recommending threat detection and mitigation to diverse audiences
CONTENT
- ML landscape & core algorithms
- Supervised learning for threat detection: Classification and regression
- Unsupervised learning and clustering techniques in cybersecurity
- Feature engineering and data preprocessing for threat detection
- Neural networks and deep learning for
- Malware detection using machine learning
- Intrusion detection systems, anomaly detection models, and network traffic analysis
- Adversarial ML: attacks & defences
- Secure ML pipeline & MLOps
- Evaluation, explainability, ethics & emerging trends
ASSESSMENT METHODS
- Lab Implementation and Technical Report – 30%
- Group Comparative Study and Design Proposal Presentation – 30%
- Research Paper and Reflective Commentary – 40%
PRESCRIBED READINGS
Check with the lecturer each semester before purchasing any texts
