Decision Tree
A model that predicts by asking a series of yes/no questions about the features — interpretable but prone to overfitting.
A decision tree predicts by splitting the data on feature thresholds in a tree of yes/no questions: each internal node tests a feature, each branch is an outcome, each leaf a prediction. It is highly interpretable — you can read the exact rules — but a single deep tree overfits easily.
Worked example: a loan model might split “income > 50k?” then “debt < 20k?” to reach an approve/deny leaf — a path you can trace and explain to a regulator. Gotcha: single trees are unstable (a small data change reshapes the tree) and overfit, which is exactly why they are almost always used in ensembles — random forests (bagging) and gradient boosting — that trade the single tree’s interpretability for far better accuracy and stability.