Core Algorithms

• Linear and logistic regression and reading their coefficients
• Decision trees, random forests and gradient boosting
• k-means and hierarchical clustering for exploratory grouping

Evaluation

• Train, validation and test splits and why the test set is touched once
• Accuracy, precision, recall, F1 and choosing by consequence
• Overfitting and underfitting diagnosed from learning curves

Data

• Missing values, encoding and scaling done inside a pipeline
• Class imbalance and its effect on naive metrics
• Leakage: the failure that makes a bad model look excellent

Practice

• Framing a problem and selecting a metric before modelling
• Building, evaluating and iterating on a baseline
• Communicating results and known limitations