Back to GalleryDecision Tree
Decision Trees recursively split the feature space into axis-aligned rectangular regions. At each node, the CART algorithm searches for the single feature and threshold that maximizes impurity reduction (Information Gain).
Key Parameters & Visual Influence:
• Max Depth: Controls tree growth. Deep trees capture complex, high-resolution nested patterns (high variance) but overfit, while shallow trees generate broad, simple axis-aligned blocks.
• Split Criterion: Gini Impurity and Entropy measure node disorder, generating different splits.
• Min Samples Split & Leaf: Require a minimum sample count to split nodes or form leaves, preventing tiny, noise-fitting sub-divisions.
Dataset Studio
CSV format: Headers in first row, last column = label/target. All values must be numeric.
Example: x1, x2, label