Random Forest

An ensemble learning method for classification, regression and other tasks that operates by constructing a multitude of decision trees.

How it works

A Random Forest trains many decision trees, each on a different random subset of the training data (bootstrap sampling) and using a random subset of features at each split. At inference, each tree produces an independent prediction, and the forest aggregates them by majority vote (classification) or averaging (regression). The randomness in data sampling and feature selection ensures the trees are diverse enough that their errors are uncorrelated, so the ensemble's error is lower than any single tree.

Why it matters

Random Forests are one of the most robust and widely used ML algorithms for structured tabular data. They handle mixed data types, missing values, and noisy features gracefully, require minimal hyperparameter tuning, and provide feature importance scores for interpretability. They are the go-to baseline for any new tabular prediction task. In many financial, healthcare, and operational prediction problems, Random Forests — or their gradient-boosted cousins like XGBoost — outperform more complex deep learning models.