About estimation of quality of clustering results via its stability
Issue title: Intelligent Computing for Pattern Recognition, Image Processing and Computer Vision
Papers from CIARP 2014, November 2-5, 2014, Puerto Vallarta, Jalisco, Mexico
Affiliations: Dorodnicyn Computing Centre, Federal Research Center ``Computer Science and Control'' of Russian Academy of Sciences, Vavilov st. 40, 119333 Moscow, Russia. Tel.: +7 499 135 4498; Fax: +7 499 135 6159; E-mail: [email protected]; http://www.ccas.ru/rvv
Abstract: We know that there are many clustering methods for the case of a known/unknown number of clusters.
Clustering is a result of fulfillment of some stopping criterion. Usually, optimisation of some quality criterion is performed or iterative processes are accomplished.
How to estimate the quality of clustering obtained by some method? Is the obtained clustering result corresponding to the objective reality or some stopping criterion of the algorithm is made and we have obtained only some partition? Here, a practical approach and the common general criteria based on
an estimation of the stability of clustering are submitted. The criterion does not use any probabilistic assumptions or distances in feature space. For some well-known clustering algorithms,
efficient methods for computing the introduced stability criteria according to the training set are obtained. Some illustrative real and artificial examples for various situations are shown.
Keywords: Clustering, stability, cluster, feature, a hierarchical grouping, variance, mean
