Further Reading and Course Map¶
Recommended readings¶
The course guidelines name three essential texts and three additional references. This book is self-contained for the course, but each of these goes deeper.
Essential
- T. M. Mitchell, Machine Learning, McGraw Hill Education, 2017.
- G. James, D. Witten, T. Hastie and R. Tibshirani, An Introduction to Statistical Learning with Applications in Python, Springer, 2023.
- E. Alpaydin, Introduction to Machine Learning, MIT Press, third edition.
Additional
- P. Flach, Machine Learning: The Art and Science of Algorithms that Make Sense of Data, Cambridge University Press, 2015.
- C. M. Bishop, Pattern Recognition and Machine Learning, Springer, 2016.
- S. Raschka, Python Machine Learning, Packt Publishing, 2019.
Where the course sends you¶
The course guidelines assign reading to each unit as follows. The number in brackets is the text above (the essential list): [1] Mitchell, [2] James, Witten, Hastie and Tibshirani, [3] Alpaydin.
| Unit | Topic | Reading |
|---|---|---|
| 1 | Introduction | Chapter 1 of [3] |
| 2 | Feature scaling, feature selection | Chapter 6 (6.1.1, 6.1.2) of [2] |
| 2 | Dimensionality reduction (PCA) | Chapter 12 (12.2) of [2] |
| 3 | Linear regression | Chapter 3 (3.1, 3.2) of [2] |
| 3 | Regularisation | Chapter 6 (6.2.1) of [2] |
| 4 | Decision trees | Chapter 3 (3.1 to 3.4) of [1] |
| 4 | Naive Bayes | Chapter 6 (6.1, 6.2, 6.7, 6.9) of [1] |
| 4 | Logistic regression | Chapter 4 (4.3.1 to 4.3.4) of [2] |
| 4 | k-nearest neighbours | Chapter 8 (8.1, 8.2) of [1] |
| 4 | Perceptron, multilayer networks | Chapter 10 (10.1, 10.2, 10.7) of [2] |
| 4 | Support vector machines | Chapter 9 (9.1 to 9.4) of [2] |
| 4 | Classification evaluation metrics | Chapter 5 (5.1) of [2]; Chapter 19 (19.7) of [3] |
| 5 | K-means and hierarchical clustering | Chapter 12 (12.4.1, 12.4.2) of [2] |
Data for practice¶
The guidelines suggest public data sets from the repositories at data.gov.in and archive.ics.uci.edu/ml/datasets.php for the practical programs.
Topic index¶
| If you want... | Go to |
|---|---|
| The T / P / E definition, supervised and unsupervised learning | Chapter 1 |
| Min-max and z-score scaling, one-hot encoding, outliers, SMOTE | Chapter 2 |
| Feature selection and PCA by hand | Chapter 3 |
| Least squares, the normal equation, gradient descent | Chapter 4 |
| Bias and variance, ridge, lasso, MSE, \(R^2\), cross-validation | Chapter 5 |
| Entropy, information gain, Gini index | Chapter 6 |
| Bayes' theorem, Laplace smoothing, Gaussian naive Bayes | Chapter 7 |
| The sigmoid, log-loss, softmax | Chapter 8 |
| Distance-based classification | Chapter 9 |
| Perceptron, forward and backward pass | Chapter 10 |
| Margins, support vectors, kernels | Chapter 11 |
| Confusion matrix, precision, recall, F1, ROC and AUC | Chapter 12 |
| Euclidean, Manhattan and cosine distance, k-means | Chapter 13 |
| Linkage, dendrograms, silhouette | Chapter 14 |