I'm trying to track down the origins of the word "cluster" and its usage in the context of cluster analysis.

Please, does anyone know when and by whom it was first used? Perhaps there was a paper or a book which coined these terms?

Alternatively, maybe there is some literature describing the beginnings of the fields which work with the term "cluster" such as machine learning or statistics of some kind?

  • It is a (very) Old English word of Germanic origin. Orig Eng clyster, Germ kluster.
    – WS2
    Sep 15, 2015 at 12:33
  • You might want to use ngrams books.google.com/ngrams to see early uses of the term.
    – Mitch
    Sep 15, 2015 at 13:57

3 Answers 3


According to Wikipedia:.

  • Cluster analysis was originated in anthropology by Driver and Kroeber in 1932 and introduced to psychology by Zubin in 1938 and Robert Tryon in 1939 and famously used by Cattell beginning in 1943 for trait theory classification in personality psychology.

and Ngram first instances of "cluster analysis" are from the 30's.

One example of early usage is from 1935:

Cluster analysis:

  • Cluster analysis was performed on matrices of Euclidean distance coefficients and Pearson product-moment correlation-coefficients with the unweighted-pair- group- method (UPGM, e.g. Sneath and Sokal, 1973, p. 114-244; Davis, 1986, p.

The following source traces its wider usage and application in the 60's:

Cluster analysis :

  • emerged as a major topic in the 1960’s and 1970’s when the monograph ’Principles of numerical taxonomy’ by Sokal and Sneath (1963) motivated world-wide research on clustering methods and thereby initiated the publication of a broad range of books such as ’Les bases de la classification automatique’ (Lerman 1970), ’Mathematical taxonomy’ (Jardine and Sibson 1971), ’Cluster analysis for applications’ (Anderberg 1973). , with the conse quence that the basic problems and methods of clustering became well-known.


The usage of the term "cluster" is quite intuitive reading its definition:

  • Statistical classification technique in which cases, data, or objects (events, people, things, etc.) are sub-divided into groups (clusters) such that the items in a cluster are very similar (but not identical) to one another and very different from the items in other clusters. It is a discovery tool that reveals associations, patterns, relationships, and structures in masses of data.




It is a (very) Old English word of Germanic origin. Orig Eng clyster, Germ kluster.

The earliest example from the OED is from the year 800 AD.

a800 Corpus Gloss. 318 (O.E.T. 45) Botrum, clystri.


Here's one of the earlier references that I can find (from 1953).

Title Cluster Analysis

Author Anatol W. Holt

Publisher Massachusetts Institute of Technology, Department of Mathematics, 1953

EDIT (Note - following a comment from JBJ it seems the following date is incorrect. I'll check and return when I have better data)

Transactions of the Institution of Mining and Metallurgy, Volume 79 - The Institution, 1893 - Metallurgy

See page B-179

  • 1
    That second source is misdated: vol. 79 of the Transactions of the Institution of Mining and Metallurgy is from 1970, not 1893 (it says so further down in the index table as well). Sep 15, 2015 at 13:29
  • @JanusBahsJacquet - Oh, thanks for that. I'll check it out and edit. Sep 15, 2015 at 13:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.