1. Course Description
This course provides a comprehensive introduction to the fundamental principles, algorithms, and practices of machine learning. Students will embark on a journey from the theoretical underpinnings of learning from data to the practical application of building, evaluating, and deploying predictive models.
The curriculum covers the entire machine learning lifecycle, including data preparation, supervised and unsupervised learning, neural networks, and the critical ethical considerations of modern AI. The course emphasizes both conceptual understanding and hands-on implementation using the Python ecosystem. It is designed for students with a background in programming and statistics who wish to build a strong foundation in the field of artificial intelligence.
2. Course Learning Outcomes
Upon successful completion of COMP328, a student will be able to:
Knowledge
- Explain the core principles of machine learning and differentiate between major learning paradigms (supervised, unsupervised, reinforcement).
- Describe the mathematical and conceptual foundations of major machine learning algorithms, including linear models, tree-based methods, and neural networks.
Skills
- Prepare and explore real-world datasets for machine learning tasks using Python libraries such as NumPy, Pandas, and Scikit-Learn.
- Implement, train, and evaluate predictive models for regression, classification, clustering, and dimensionality reduction problems.
- Interpret model results using appropriate evaluation metrics (e.g., accuracy, precision, recall, F1-score, ROC curves) and diagnose issues like bias and variance.
- Design and execute a complete machine learning project from problem definition to final presentation.
Professional Competencies
- Assess the ethical implications of machine learning models, including identifying potential sources of bias and unfairness.
- Communicate machine learning findings and technical concepts clearly and effectively to both technical and non-technical audiences.
3. Required Materials
Primary Textbook
There is no single required textbook. All necessary reading materials, articles, and documentation will be provided within the course modules.
Software
- Python 3.9+
- Jupyter Notebooks / JupyterLab
- Core Python libraries: NumPy, Pandas, Matplotlib, Scikit-Learn
Hardware
- A modern computer with a stable internet connection capable of running a web browser and the Python environment.
4. Course Structure & Content Outline
The course is organized into six modules comprising 20 units. Each unit contains a landing page, multiple topic pages with readings and activities, and a unit quiz.
Module 1 · Foundations of Machine Learning
Conceptual grounding before you build anything: what learning from data actually means.
- Unit 1Introduction to Machine Learning — history of AI/ML, ML lifecycle, applications, limitations.
- Unit 2Data Fundamentals — structured vs. unstructured data, features, labels, data quality, preprocessing, dataset splitting.
- Unit 3Python ML Ecosystem — Jupyter, NumPy, Pandas, Matplotlib, Scikit-Learn.
Module 2 · Supervised Learning
Predict continuous and categorical outcomes, and learn when a model is genuinely useful.
- Unit 4Linear Regression — prediction problems, regression models, cost functions, model fitting.
- Unit 5Logistic Regression — classification, probability, the sigmoid function, decision boundaries.
- Unit 6Model Evaluation — accuracy, precision, recall, F1-score, ROC curves, cross-validation, bias–variance tradeoff.
Module 3 · Classical Machine Learning Algorithms
The workhorse algorithms of applied ML — interpretable, fast, and still widely deployed.
Module 4 · Unsupervised Learning
Find structure without labels: grouping, compression, association, and the unusual.
- Unit 10Clustering — K-Means, hierarchical clustering, cluster quality.
- Unit 11Dimensionality Reduction — curse of dimensionality, PCA, visualization.
- Unit 12Association Learning — market basket analysis, Apriori algorithm, rules and confidence.
- Unit 13Anomaly Detection — outliers, fraud detection, intrusion detection.
Module 5 · Neural Networks and Deep Learning
From a single artificial neuron to convolutional architectures that see.
Module 6 · Responsible and Applied Machine Learning
Ship it, monitor it, and govern it — the professional realities of ML in production.
- Unit 17Model Deployment — production pipelines, APIs, model monitoring, MLOps overview.
- Unit 18Ethics — bias, privacy, transparency, accountability.
- Unit 19Governance and Fairness — algorithmic fairness, responsible AI, emerging regulations.
- Unit 20Course Integration — comprehensive review, exam preparation, capstone preparation.
5. Assessment & Grading
Your final grade will be determined by your performance on a combination of quizzes, assignments, and a final capstone project.
Unit Quizzes
Twenty end-of-unit quizzes covering the core concepts from each unit’s readings and activities.
Assignments
Five hands-on projects applying ML techniques to real-world datasets, from data exploration through to neural network modeling.
Capstone Project
A complete ML lifecycle project integrating problem selection, data work, modeling, evaluation, ethics, and a final presentation.
Participation
Constructive engagement in discussion forums and peer review activities throughout the term.
Assignment Breakdown
- Assignment 1 Data Exploration and Preparation Open
- Assignment 2 Regression Modeling and Evaluation Open
- Assignment 3 Classification Analysis Open
- Assignment 4 Unsupervised Learning Techniques Open
- Assignment 5 Neural Network Modeling Open
- Capstone End-to-end machine learning project Open
Grading Scale
| Letter Grade | Percentage Range |
|---|---|
| A+ | 95–100% |
| A | 90–94% |
| A− | 85–89% |
| B+ | 80–84% |
| B | 75–79% |
| B− | 70–74% |
| C+ | 65–69% |
| C | 60–64% |
| D | 50–59% |
| F | Below 50% |
6. Course Policies
Late Submissions
Late assignments will be penalized by 10% per day, up to a maximum of three days. Extensions may be granted for extenuating circumstances with prior approval from the instructor.
Academic Integrity
All work submitted in this course must be your own. Plagiarism, cheating, and any other form of academic dishonesty will be taken seriously and will result in a failing grade for the assignment and potentially the course. Please refer to the university’s academic integrity policy for more details.
Communication
All official course communication will be conducted through the course announcement board and email. Students are expected to check these channels regularly. For personal questions, please use email or attend office hours.
Accessibility
This course is committed to providing an accessible learning environment. If you require accommodations, please contact the university’s disability services office and inform the instructor as early as possible.
7. Course Schedule (Tentative)
Note
This schedule is tentative and subject to change. Any updates will be announced through the course announcement board.
| Week | Module | Topics & Units Covered | Assessments Due |
|---|---|---|---|
| 1 | 1 | Course Intro, Unit 1: Introduction to ML | — |
| 2 | 1 | Unit 2: Data Fundamentals | — |
| 3 | 1 | Unit 3: Python ML Ecosystem | — |
| 4 | 2 | Unit 4: Linear Regression | Assignment 1 |
| 5 | 2 | Unit 5: Logistic Regression | — |
| 6 | 2 | Unit 6: Model Evaluation | Assignment 2 |
| 7 | 3 | Unit 7: Decision Trees | — |
| 8 | 3 | Unit 8: Random Forests, Unit 9: Support Vector Machines | — |
| 9 | 4 | Unit 10: Clustering, Unit 11: Dimensionality Reduction | Assignment 3 |
| 10 | 4 | Unit 12: Association Learning, Unit 13: Anomaly Detection | — |
| 11 | 5 | Unit 14: Neural Networks, Unit 15: Deep Learning | Assignment 4 |
| 12 | 5 | Unit 16: Convolutional Neural Networks | — |
| 13 | 6 | Unit 17: Model Deployment, Unit 18: Ethics | Assignment 5 |
| 14 | 6 | Unit 19: Governance and Fairness, Unit 20: Course Integration | Capstone Project Due |
| 15 | — | Final Exam Period (No Exam — Capstone Presentations) | — |