A decision tree classifies by asking yes/no questions about one feature at a time (“is x < 0.4?”), each answer leading to the next question until a leaf names a class. It is grown greedily: at each region, try every question and keep the one whose two halves are purest.