Random Forests
Leo Breiman published "Random Forests" in Machine Learning in October 2001. The paper describes random forests as combinations of tree predictors in which each tree depends on a randomly sampled vector, and analyzes forest generalization error in terms of individual-tree strength and correlation.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented machine-learning milestone for random-forest ensemble methods, recorded from the original Springer publisher record, Breiman's author-hosted paper copy, and a University of California, Berkeley technical-report record documenting closely related earlier random-forest work.
People, organizations & technologies
PEOPLE
Leo Breiman
TECHNOLOGIES
Random forests, Ensemble learning, Tree predictors
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-14T16:06:05+08:00
Springer records the Machine Learning article as issue date October 2001, volume 45, pages 5-32, DOI 10.1023/A:1010933404324. Breiman's author-hosted January 2001 paper copy independently exposes the paper text and supports the method description, including randomly sampled vectors, generalization error, tree strength/correlation, random feature selection, variable importance, and regression use; its January manuscript date is not treated as the journal-publication month. UC Berkeley separately preserves Breiman's September 1999 technical report "Random Forests--Random Features". LINEAiGE records only the October 2001 journal-publication milestone here and does not infer a universal invention date, development geography, ancestry, influence, causality, or relationships.
Sources
Random Forests
Leo Breiman / University of California, Berkeley
Random Forests--Random Features
University of California, Berkeley Department of Statistics
Trace the chronology
Record identity
LINEAiGE IDrandom-forests-2001