BernoulliMix program package provides tools to work with finite mixture models of multivariate Bernoulli distributions, also known as Bernoulli mixtures. The program package can be used for probabilistic modeling of 0-1 data. The target audience includes researchers, teachers, and students in machine learning and data mining.
BernoulliMix program package is used on our machine learning course T-61.5140 Machine Learning: Advanced Probabilistic methods at the Helsinki University of Technology (TKK) in Finland as the core of the term project. Instead of making students program most of the solutions by themselves, we give them a ready implementation and make them design experimental runs, produce results and think of their meaning. I gave a talk about experiences in teaching at the Workshop on Teaching Machine Learning in Saint Etienne, France in May 2008. You may want to take a look at the slides of the talk Learning mixture models — courseware for finite mixture models of multivariate Bernoulli distributions or the paper Learning mixture models itself. The current version (v. 1.11) of the documentation included in the package is also separately available in (PostScript, 329k), (PDF, 194k) formats.
You can download the BernoulliMix package by moving to the download page.
© 2002-2009 Jaakko Hollmén, Helsinki University of Technology, Finland. BernoulliMix program package is licensed under GNU General Public License 3.0 and GNU Free Documentation License 2.1. You can freely observe the source code of all programs in BernoulliMix programming package in C programming language, learn from the programs, and make modifications under the terms of the license. If you think your contribution is useful to other users of BernoulliMix, send your contribution to the author: Jaakko Hollmén. Also, suggestions for improvements and general feedback is encouraged.