Rankcluster: An R Package for Clustering Multivariate Partial Rankings

Authors

Julien Jacques, University Lille I
Quentin Grimonprez, MODAL team
Christophe Biernacki, University Lille I

Date of this Version

6-2014

Document Type

Article

Citation

The R Journal (June 2014) 6(1); Editor: Deepayan Sarkar

Comments

Copyright 2014, The R Foundation. Open access material. License: CC BY 3.0 Unported

Abstract

The Rankcluster package is the first R package proposing both modeling and clustering tools for ranking data, potentially multivariate and partial. Ranking data are modeled by the Insertion Sorting Rank (ISR) model, which is a meaningful model parametrized by a central ranking and a dispersion parameter. A conditional independence assumption allows multivariate rankings to be taken into account, and clustering is performed by means of mixtures of multivariate ISR models. The parameters of the cluster (central rankings and dispersion parameters) help the practitioners to interpret the clustering. Moreover, the Rankcluster package provides an estimate of the missing ranking positions when rankings are partial. After an overview of the mixture of multivariate ISR models, the Rankcluster package is described and its use is illustrated through the analysis of two real datasets