Welcome to NT120
This course provides an introductory foundation in Artificial Intelligence and Machine Learning, with the emphasis placed on problem formulation, data preparation, baseline model development and result evaluation in computer networking and cybersecurity contexts. Students work through guided exercises on small datasets derived from network traffic, logs and security events, and apply selected baseline classification and anomaly-detection methods. k-Nearest Neighbours, Naive Bayes and Support Vector Machines are introduced briefly so that students recognise their operating principles and appropriate use cases. Deep learning, Transformers and large language models, Federated Learning, Reinforcement Learning and Deep Reinforcement Learning are presented as future study directions connected to advanced courses and representative research problems. The course also addresses adversarial AI, privacy, bias and the responsible use of AI in networking and cybersecurity.
Course Information
Course Code
NT120
Credits
2 lecture credits; no separate practice credits
Independent Study
30 hours
Prior Courses
Introduction to Computer Networks; Introduction to Programming
Learning Outcomes
CLO G1.1: Concepts
Explain AI/ML concepts, supervised and unsupervised learning, and representative applications. Level: Teach.
CLO G1.2: Workflow
Describe a basic workflow from problem formulation and data preparation through to model evaluation. Level: Teach.
CLO G2.1: Data
Preprocess and explore a small networking or cybersecurity dataset using a guided notebook. Level: Utilise.
CLO G2.2: Modelling
Train a baseline classification or anomaly-detection model using available tools. Level: Utilise.
CLO G3.1: Evaluation
Interpret a confusion matrix, accuracy, precision, recall and F1-score, together with the limitations of the results. Level: Utilise.
CLO G4.1: Risk and Responsibility
Analyse risks from adversarial AI, generative AI, privacy and bias, and recognise advanced study directions. Level: Teach.
Instructors
Principal Textbooks
-
BOOK
Machine Learning for Computer and Cyber Security: Principles, Algorithms, and Practices
Gupta & Sheng (CRC Press, 2019)
-
BOOK
Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow (3rd ed.)
Géron (O'Reilly, 2023)
-
BOOK
Machine Learning and Security: Protecting Systems with Data and Algorithms
Chio & Freeman (O'Reilly, 2018)
References
-
BOOK
Deep Learning
Goodfellow, Bengio & Courville (MIT Press, 2016) • deeplearningbook.org
-
BOOK
Introduction to Machine Learning with Applications in Information Security (2nd ed.)
Stamp (Chapman & Hall/CRC, 2022)
-
BOOK
Hands-On Artificial Intelligence for Cybersecurity
Sumeet & Shrey (Packt Publishing, 2024)
-
WEB
MITRE ATT&CK Knowledge Base
attack.mitre.org
-
WEB
MITRE ATLAS: Adversarial Threat Landscape for AI Systems
atlas.mitre.org
-
STD
Artificial Intelligence Risk Management Framework (AI RMF 1.0)
Tabassi, NIST AI 100-1 (2023)
Toolchain
Environment
Data and Analysis
Machine Learning
Demonstrations Only
No student installation required
Guided Practice
The course carries no separate practice credits. Notebook work is completed in class and at home, and the evidence is assessed within component A1.
Mini-Lab Topics
Suggested small exercises for the integrated mini-lab and self-study. Filter by theme.
Assessment
Topic Distribution
| Component | Outcomes | Weight |
|---|---|---|
| A1. Continuous assessment (quizzes, exercises, notebooks, discussion) | G1.1, G1.2, G2.1, G2.2 | 15% |
| A2. Midterm examination (written / multiple choice) | G1.1, G1.2, G3.1 | 35% |
| A3. Practical work (embedded in A1) | G2.1, G2.2 | 0% |
| A4. Final examination (multiple choice and written responses) | G2.2, G3.1, G4.1 | 50% |
Continuous Assessment
- Concepts and workflow explained accurately and related to networking or cybersecurity problems
- Notebook completed correctly, with reasonable results and a clear explanation
- Exercises, discussion and self-study completed, with active participation and clear reasoning
Scored on a five-point scale from Poor (0) to Excellent (4).
Midterm Examination
- Pipeline stages and their relationships explained accurately
- Baseline model selected and justified appropriately for the problem
- Metrics calculated and interpreted correctly, with limitations identified
Scored on a five-point scale from Poor (0) to Excellent (4).
Final Examination
- Method selected with convincing justification
- Results analysed accurately and in a balanced way, with limitations identified
- Risks, impacts and suitable mitigations for responsible AI use analysed clearly
Scored on a five-point scale from Poor (0) to Excellent (4).
Course Requirements and Expectations
Attendance and Submissions
- Attendance at a minimum of 70% of scheduled sessions is required
- Late assignments and assessments are not graded unless authorised under institutional regulations
- Notebook work forms the practical evidence for component A1, so it must be submitted on schedule
Academic Integrity
- Academic misconduct is handled in accordance with University regulations
- Students remain responsible for verifying any AI-generated output they submit, in line with the course material on responsible AI use