Nice work! FWIIW, Mixture, Naive Bayes and semisupervised versions of each with the EM-algo can often be cooked up from the same raw materials. And multivariate/Bernoulli, I think can be derived from the mean term in the Gaussian example. Multivariate and Bernoulli should probably be done with sparse arrays, but with these 2-3 things, supervised, unsupervised and semi-sup, you can solve an awful lot of data science problems!
-SL ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
