Genie (R Package)

This project has been superseded by genieclust, which features a faster and more feature-rich implementation of Genie (now also available for both R and Python).

A Fast and Robust Hierarchical Clustering Algorithm

The time needed to apply a hierarchical clustering algorithm is most often dominated by the number of computations of a pairwise dissimilarity measure. Such a constraint, for larger data sets, puts the use of all the classical linkage criteria at a disadvantage, with the exception of the single linkage one. However, it is known that the single linkage clustering algorithm is very sensitive to outliers, produces highly skewed dendrograms and therefore usually does not reflect the true underlying structure of analysed data - unless the clusters are well-separated. To overcome its limitations, we proposed a hierarchical clustering linkage criterion called Genie. Namely, our algorithm links two clusters in such a way that the Gini measure of inequity of the cluster sizes does not exceed a given threshold. This method most often outperforms the Ward or average linkage in terms of the clustering quality on benchmark data. At the same time, Genie retains the high speed of the single linkage approach, therefore it is also suitable for analysing larger data sets. The algorithm is easily parallelizable and thus may be run on multiple threads to speed up its execution further on. Its memory overhead is small: there is no need to precompute the complete distance matrix to perform the computations in order to obtain a desired clustering.

A detailed description of the algorithm can be found in:

Gagolewski M., Bartoszuk M., Cena A., Genie: A new, fast, and outlier-resistant hierarchical clustering algorithm, Information Sciences, 2016, doi:10.1016/j.ins.2016.05.003.

Authors: Marek Gagolewski, Maciej Bartoszuk, and Anna Cena

CRAN entry: http://cran.r-project.org/web/packages/genie/

See also: http://genieclust.gagolewski.com/