Research Note

A fast version of the k-means classification algorithm for astronomical applications

¹ Instituto de Física de Cantabria, Avenida de los Castros, s/n, 39005 Santander, Spain
e-mail: ordovas@ifca.unican.es
² Departamento de Astrofísica, Universidad de La Laguna, 38205, La Laguna, Tenerife, Spain
³ Instituto de Astrofísica de Canarias, 38205, La Laguna, Tenerife, Spain
e-mail: jos@iac.es

Received: 14 March 2014
Accepted: 28 March 2014

Abstract

Context. K-means is a clustering algorithm that has been used to classify large datasets in astronomical databases. It is an unsupervised method, able to cope very different types of problems.

Aims. We check whether a variant of the algorithm called single pass k-means can be used as a fast alternative to the traditional k-means.

Methods. The execution time of the two algorithms are compared when classifying subsets drawn from the SDSS-DR7 catalog of galaxy spectra.

Results. Single-pass k-means turn out to be between 20% and 40% faster than k-means and provide statistically equivalent classifications. This conclusion can be scaled up to other larger databases because the execution time of both algorithms increases linearly with the number of objects.

Conclusions. Single-pass k-means can be safely used as a fast alternative to k-means.