Index

MIXMOD 3rd Mixmod meeting : call for presentation See Roadmap Patch for Scilab 5.1. (and further)

MIXMOD (MIXture MODelling) is a software for density estimation, clustering or discriminant analysis problems.

A large variety of algorithms is proposed (EM, Classification EM (CEM), Stochastic EM (SEM)) to maximize likelihood or completed likelihood.

MIXMOD can be used on quantitative data (thanks to multivariate Gaussian mixtures models) or qualitative data (thanks to multivariate Multinomial mixtures models).

Moreover, some criteria (BIC, ICL, NEC and CV) allow to choose the best model among parsimonious models (22 in quantitative context (8 are specific for High Dimension data) and 5 in qualitative context).

Written in C++, MIXMOD is interfaced with Scilab and Matlab and distributed under GNU General Public License(GPL).

MIXMOD is been developed jointly by :

• Florent Langrognet - Laboratory of Mathematics of Besançon
• Gilles Celeux - Inria (Select Project)
• Christophe Biernacki - Laboratory of Mathematics of Lille
• Gérard Govaert - Laboratory Heudiasyc (Compiègne)