Computer Science · Theme A: Concepts of computer science

A4.3 — Machine learning approaches

Computer Science · SL / HL · syllabus-mapped notes

1. CS_A4.3.1

   Linear regression for continuous outcomes

   Predictor and response variables, slope and intercept, and fit measured by r squared.

2. CS_A4.3.2

   Classification: K-NN and decision trees

   Predicting discrete categories from labelled data with K-NN and decision trees.

3. CS_A4.3.3

   Evaluating models and hyperparameter tuning

   Accuracy, precision, recall, F1, tuning, and overfitting versus underfitting.

4. CS_A4.3.4

   Clustering in unsupervised learning

   Grouping data by similarities in its features.

5. CS_A4.3.5

   Association-rule learning

   Uncovering relations between attributes in large data sets.

6. CS_A4.3.6

   Reinforcement learning

   An agent learning from its environment via cumulative reward.

7. CS_A4.3.7

   Genetic algorithms

   Population, fitness, selection, crossover and mutation applied to optimization.

8. CS_A4.3.8

   ANNs: perceptron and multi-layer networks

   The structure and function of artificial neural networks, with perceptron and MLP sketches.

9. CS_A4.3.9

   Convolutional neural networks (CNNs)

   Learning spatial hierarchies of features in images through its layers.

10. CS_A4.3.10

    Model selection and comparison

    Why different algorithms suit different data and problems.