Random Forest Classifier

Harvard EPS-210 | Interactive tutorial — Explore ensemble learning with bagging and random feature selection

Ensemble Decision (Majority Vote)

Class 0
Class 1
Query

Individual Tree Boundaries

Each color = one tree's boundary

Bootstrap Samples (Bagging)

Feature Importance

Add Data Points

Forest Statistics

0
Trees
--
OOB Acc
--
Train Acc

Forest Parameters

Number of Trees 10
Max Depth 4
Bootstrap Sample % 80%

Model Statistics

Total Points
0
Avg Depth
--
Class 0
0
Class 1
0
Hover to see tree votes
--
--

Quick Examples

How it works: Random Forest builds multiple decision trees on bootstrap samples (bagging) with random feature subsets. Final prediction is by majority vote, reducing overfitting and variance.