Some methods for classification and analysis of multivariate observations
J. MacQueen published 'Some methods for classification and analysis of multivariate observations' in the Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability in 1967. The paper describes a process for partitioning an N-dimensional population into k sets and explicitly calls the process 'k-means.'
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented 1967 publication milestone for the k-means method, represented strictly as a publication event rather than an invention or priority claim.
People, organizations & technologies
PEOPLE
J. MacQueen
TECHNOLOGIES
k-means
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-15T05:03:30+08:00
The University of California, Berkeley digital archive preserves MacQueen's original proceedings paper and its text explicitly states that the described partitioning process is called 'k-means.' Independent National Diet Library / CiNii bibliographic metadata records the publication year as 1967, in the Proceedings of the Fifth Berkeley Symposium, volume 1, pages 281-297, University of California Press. The Berkeley archive also exposes the symposium session dates spanning 1965-1966, which are not treated here as the publication date. This candidate therefore preserves year-only publication precision and does not assert an invention date, first-ever claim, development geography, ancestry, influence, causality, or relationship edge.
Sources
Some methods for classification and analysis of multivariate observations
University of California, Berkeley Digital Collections
Some methods for classification and analysis of multivariate observations
National Diet Library Search / CiNii Research
Trace the chronology
Record identity
LINEAiGE IDmacqueen-k-means-1967