About Course
Machine learning is not just about importing a library, training a model and hoping for the best. This course helps you understand what is happening underneath the code, so you can choose, evaluate and improve models with confidence.
You will build a strong foundation in the statistical and mathematical ideas behind supervised machine learning, covering regression, classification, validation, model selection, regularisation, PCA and decision trees. We explore not only how popular algorithms work, but also when to use them, how to assess whether they are performing well, and where they can fail in real-world settings.
Across 79 lessons and over 20 hours of content, you will learn topics including linear and logistic regression, k-nearest neighbours, support vector machines, cross-validation, bootstrapping, bias-variance trade-offs, AIC and BIC, Ridge and Lasso regression, and evaluation metrics such as accuracy, precision, recall and ROC curves. Python code discussions and suggested exercises help connect the theory to practical application.
This course is ideal for aspiring data scientists, software engineers moving into AI, students, and professionals who want to go beyond using machine-learning tools as black boxes.
What Will You Learn?
- - Machine Learning Principles
- - The principles behind Machine Learning algorithms (not just the codes!)
- - Regression (Linear Regression, Multiple Linear Regression, Polynomial Regression, and Support Vector Regression)
- - Classification (Logistic Regression, k-Nearest Neighbours, Trees, and Support Vector Machines)
- - Other principles such as Cross Validation, AIC, BIC, and choosing the right metrics for your algorithm
Course Content
Introduction to Machine Learning
-
Introduction
07:37
Introduction to Statistical Learning
-
Modelling for prediction versus modelling for interference
29:47 -
Parametric versus Non Parametric Method
20:53 -
Model Prediction Accuracy and Model Interpretability Trade Off
13:36 -
Supervised versus Unsupervised Learning
12:22 -
Regression versus Classification
11:58 -
Assessing Model Accuracy – Measury of Quality of Fit
31:30 -
Bias Variance Trade Off
22:31 -
Assessing Model Accuracy Classification Setting
10:47 -
Classification Example K Nearest Neighbours KNN
19:56 -
Confidence Intervals for Coefficient Estimates: Simple Linear Regression
10:01 -
Hypothesis Test of Coefficient Estimates: Simple Linear Regression
22:35 -
Accuracy of Coefficient Estimates: Simple Linear Regression
29:43 -
Estimating Model Coefficients: Simple Linear Regression
28:46
Linear Regression
-
Introduction to Linear Regression
11:08 -
Assessing Simple Linear Model Accuracy
01:56 -
Residual Standard Error (RSE)
08:34 -
R- Squared Statistic
17:04 -
Multiple Linear Regression
12:17 -
Estimating Multiple Linear Regression Coefficients
17:37 -
Question 1 – Is There a Relationship Between Response and Predictors
15:17 -
Question 2 – Variable Selection
23:38 -
Question 3 – Model Accuracy
10:24 -
Dealing with Qualitative Variables
25:14 -
Including Interaction Terms in the Model (Non additive Linear Models)
25:12 -
Including Non Linear Terms in Linear Models
11:14 -
Problem #1 – Non linearity of the Data
05:42 -
Problem #2 – Correlation of the Error Terms
05:15 -
Problem #3 – Nonconstant Variance of Error Terms
04:22 -
Problem #4 – Outliers
07:30 -
Problem #5 – High Leverage Points
06:06 -
Problem #6 – Collinearity
16:01 -
Python Code Discussion – Regression
32:30
Classification
-
Introduction to Classification
08:06 -
Why Linear Regression Will Not Work Perfectly In Classification
10:15 -
Introduction to Logistic Regression
08:38 -
The Logistic Model
21:59 -
Estimating Logistic Regression Coefficients Maximum Likelihood Method
17:08 -
Making Predictions with Logistic Regression
11:14 -
Multiple Logistic Regression
-
Introduction to Linear Discriminant Analysis (LDA)
07:08 -
Bayes’ Theorem of Classification
14:09 -
Linear Discriminant Analysis One Predictor
24:30 -
Linear Discriminant Analysis with More Predictors
20:16 -
The Confusion Matrix, Sensitity and Specificity
33:27 -
The ROC Curve
14:54 -
Quadratic Discriminant Analysis
08:23 -
Introduction to Resampling Methods
05:32 -
Method 1 – Validation Set Approach
08:04 -
Method 2 – Leave One Out Cross Validation
12:50 -
Method 3 – k-Fold Cross Validation
12:52 -
Cross Validation for Classification Settings
04:07 -
Python Code Discussion – Classification
21:51
Validation and the Bootstrap Method
-
Introduction to Resampling Methods
05:32 -
Method 1 – Validation Set Approach
08:04 -
Method 2 – Leave One Out Cross Validation
12:50 -
Method 3 – k-Fold Cross Validation
12:52 -
Cross Validation for Classification Settings
04:07 -
The Bootstrap Method
-
Python Code Discussion – Classification
20:51
Linear Model Selection and Regularisation
-
Introduction to Model Selection and Regularisation
18:07 -
Best Subset Selection Method
29:14 -
Forward Stepwise Selection Method
27:56 -
Backward Stepwise Selection Method
15:41 -
Model Selection Choosing the Optimal Model
04:59 -
Cp Estimate for Test Error
08:28 -
Akaike Information Criterion (AIC) Estimate for Test Error
03:40 -
Bayesian Information Criterion (BIC) Estimate for Test Error
05:26 -
Adjusted R-Squared Estimate for Test Error
08:29 -
Validation Set and Cross Validation Estimates for Test Error
05:13 -
Introduction to Shrinkage Methods
04:23 -
Introduction to Ridge Rigression
14:05 -
Ridge Regression Example
26:06 -
The Lasso
05:57 -
The Lasso Example
09:14 -
Lasso & Ridge Rigression using Mathematical Optimisation
29:44 -
Introduction to Dimension Reduction Methods
13:48 -
Principal Component Analysis
16:45 -
Principal Component Analysis Example
38:28 -
Scree Plots and How PCA Actually Reduces Dimensions
11:15 -
Principal Component Regression
15:42
Tree Based Methods
-
Introduction to Tree Based Methods
04:57 -
Regression Decision Trees
17:40 -
Predicting Using a Regression Decision Tree
25:25 -
Pruning Decision Trees
18:47
Student Ratings & Reviews
