Ahn, Y.-Y. | bioRxiv Science

bioRxiv2018

CluSim: a Python package for the comparison of clusterings and dendrograms

Quantifying the similarity of clusterings is a fundamental step in data analysis. Clustering similarity is the basis for method evaluation, consensus clustering, and tracking the temporal evolution of clusters, among many other tasks. Here we provide CluSim, a comprehensive Python package for the comparison of partitions, overlapping clusterings, and hierarchical clusterings (dendrograms) with more than 20 similarity measures. The CluSim package provides both analytic and empirical methods for assessing the similarity of clusterings in the context of a random model, and provides the novel element-centric approaches for clustering similarity measure that we introduced recently. We illustrate the use of the package through two examples: an evaluation of the clustering of Gene Expression data in the context of different random models, and detailed analysis of model incongruence using element-centric comparisons between a set of phylogentic trees (dendrograms).\n\nAvailability and implementationThe CluSim Python package and accompanying jupyter notebook is available at https://github.com/Hoosier-Clusters/clusim with the MIT open source licence.\n\nContactajgates42@gmail.com or yyahn@iu.edu

bioinformatics↗

bioRxiv2017

The Impact of Random Models on Clustering Similarity

Clustering is a central approach for unsupervised learning. After clustering is applied, the most fundamental analysis is to quantitatively compare clusterings. Such comparisons are crucial for the evaluation of clustering methods as well as other tasks such as consensus clustering. It is often argued that, in order to establish a baseline, clustering similarity should be assessed in the context of a random ensemble of clusterings. The prevailing assumption for the random clustering ensemble is the permutation model in which the number and sizes of clusters are fixed. However, this assumption does not necessarily hold in practice; for example, multiple runs of K-means clustering returns clusterings with a fixed number of clusters, while the cluster size distribution varies greatly. Here, we derive corrected variants of two clustering similarity measures (the Rand index and Mutual Information) in the context of two random clustering ensembles in which the number and sizes of clusters vary. In addition, we study the impact of one-sided comparisons in the scenario with a reference clustering. The consequences of different random models are illustrated using synthetic examples, handwriting recognition, and gene expression data. We demonstrate that the choice of random model can have a drastic impact on the ranking of similar clustering pairs, and the evaluation of a clustering method with respect to a random baseline; thus, the choice of random clustering model should be carefully justified.

bioinformatics↗