Section: New Software and Platforms

MixAll

Functional Description

MixAll (Clustering using Mixture Models) is a model-based clustering package for modelling mixed data sets. It has been engineered around the idea of easy and quick integration of any kind of mixture models for any kind of data, under the conditional independence assumption. Currently five models (Gaussian mixtures, categorical mixtures, Poisson mixtures, Gamma mixtures and kernel mixtures) are implemented. MixAll has the ability to natively manage completely missing values when assumed as random. MixAll is used as an R package, but its internals are coded in C++ as part of the STK++ library (www.stkpp.org ) for faster computation.

• Participant: Serge Iovleff

• Contact: Serge Iovleff

• URL: https://cran.r-project.org/web/packages/MixAll/