Machine Learning

From cleaning a dataset to regression, trees, Bayes, SVMs, neural networks, clustering, ensembles and fairness — the core algorithms of machine learning, taught by worked example.

Lecture notes by Dr. Abdulkarim Albanna · Course 606361 · References: James et al. (ISLP), Murphy (PML), Barocas, Hardt & Narayanan.

7 / 12

Modules available

Python

scikit-learn labs

Case Study

Cost

Course Modules

The modules follow the weekly plan of the course syllabus; new modules are added as the semester goes on. Each one is a written tutorial with figures, a worked example with real numbers, a Python lab, and exercises with full solutions.

Introduction to ML

AI vs ML vs DL, T–P–E, supervised / unsupervised / reinforcement learning, bias–variance.

Weeks 1–3

Data Preprocessing

Attribute types, cleaning, missing values, binning, normalization, reduction and sampling.

Weeks 1–3

Linear Regression

Least squares, gradient descent, simple & multiple regression, Ridge and Lasso.

Week 4

Logistic Regression

Binary classification, the sigmoid, odds & logit, maximum likelihood, the decision threshold.

Weeks 4–5

Performance Measures

MAE, MSE, RMSE, R²; confusion matrix, precision, recall, F1, ROC and AUC.

Week 5

Decision Trees (ID3)

Entropy, information gain, building the tree, continuous attributes, overfitting & pruning.

Week 6

Naïve Bayes

Bayes' theorem, prior / likelihood / posterior, NB variants, Laplace smoothing.

Week 7

Support Vector Machines

Maximum margin, the optimization objective, soft margins, support vectors and the kernel trick.

Coming soon · Week 8

Artificial Neural Networks

Neuron model, activations, perceptron, ADALINE, forward propagation, loss and the MLP.

Coming soon · Weeks 9–10

K-Means Clustering

Distance, the k-means objective, the algorithm step by step, and the elbow method.

Coming soon · Week 11

Ensemble Learning

Hard & soft voting, bagging and random forest, AdaBoost, gradient boosting, XGBoost.

Coming soon · Week 12

Algorithmic Fairness

Sources of bias, fairness definitions, fairness-aware ML and the FATE framework.

Coming soon · Week 13

Notebooks & Lab Files

The Python notebooks and spreadsheet used in class. Download them and open in Google Colab (File → Upload notebook) or Jupyter.

Linear Regression — Part 1

Fit a line to house size vs. price with scikit-learn, plot it, and read off the equation.

.ipynb

Linear Regression — Part 2

More examples: synthetic data, California housing, study hours vs. exam score; R² and MSE.

.ipynb

LR Calculator

The least-squares worked example as a spreadsheet: every column of the hand calculation.

.xlsx

Logistic Regression Example

The sigmoid, a decision boundary, log-loss, and the full set of classification metrics.

.ipynb

Case Study Analysis

In week 14 you apply the course to a real dataset: frame the problem, prepare the data, compare several algorithms with the right metrics, discuss fairness, and present a written report (10% of the grade).

Case study guide — coming soon

Textbooks & References

T1 G. James, D. Witten, T. Hastie, R. Tibshirani — An Introduction to Statistical Learning with Applications in Python, Springer, 2023.
T2 K. P. Murphy — Probabilistic Machine Learning: An Introduction, MIT Press, 2022.
T3 S. Barocas, M. Hardt, A. Narayanan — Fairness and Machine Learning: Limitations and Opportunities, MIT Press, 2023.
R1 C. M. Bishop, H. Bishop — Deep Learning: Foundations and Concepts, Springer, 2024.
R2 M. Hardt, B. Recht — Patterns, Predictions, and Actions, Princeton University Press, 2022.
R3 G. James et al. — An Introduction to Statistical Learning with Applications in R, 2nd ed., Springer, 2021.
R4 K. P. Murphy — Probabilistic Machine Learning: Advanced Topics, MIT Press, 2023.
R5 A. Géron — Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 3rd ed., O'Reilly, 2022.

Next: Deep Learning

Finished Module 9 and want to go deeper? The Deep Learning course continues from the multi-layer perceptron: backpropagation in full, optimizers, CNNs and RNNs.

Open Deep Learning

About This Course

These notes cover the Machine Learning course (606361) and are built from Dr. Abdulkarim Albanna's lecture slides, completed with the course textbooks. The course takes a theoretical and a practical approach: each algorithm is explained in plain English, worked through by hand with real numbers, then implemented in Python with scikit-learn. The aims are to understand the mathematics behind the algorithms, to compare them and choose the right one for a problem, and to use them responsibly — which is why the course ends with algorithmic fairness.