Random Forests--Random Features
Leo Breiman's University of California, Berkeley Statistics technical report 'Random Forests--Random Features' is dated September 1, 1999. The report describes random forests as combinations of tree predictors in which each tree depends on an independently sampled random vector, and discusses random feature selection for tree splitting.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented research-publication milestone for random forests, represented strictly as the dated Berkeley technical-report record rather than as a universal invention or release date.
People, organizations & technologies
PEOPLE
Leo Breiman
ORGANIZATIONS
University of California, Berkeley
TECHNOLOGIES
Random forests, Decision trees, Ensemble learning, Random feature selection
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-12T08:07:47+08:00
The UC Berkeley Statistics Department technical-report record directly identifies Leo Breiman, report number 567, the title 'Random Forests--Random Features', and September 1, 1999, and its abstract supports the recorded technical description. Springer Nature's later 2001 journal article independently corroborates Breiman's authorship, the random-forests formulation, and his Statistics Department, University of California, Berkeley affiliation. September 1, 1999 is attached only to this technical-report milestone; it is not asserted as a universal invention, software-release, or system-birth date. University of California, Berkeley is recorded as report/author institutional context, not as inferred development geography. No ancestry, influence, causality, or relationship edge is inferred.
Sources
Random Forests--Random Features
Department of Statistics, University of California, Berkeley
PUBLISHED
1999-09-01
Trace the chronology
Record identity
LINEAiGE IDrandom-forests-random-features-1999