MGLM

Multivariate count data abound in modern application areas such as genomics, sports, imaging analysis, and text mining. Because of the complex structure of the data, the popular multinomial logit model might not be sufficient. The R package MGLM and matlab toolbox mglm provide functions that fit distributions, run regression and penalized regression on the multivariate count data. Implementations of multinomial logit model, Dirichlet multinomial model, generalized Dirichlet multinomial model, and negative multinomial models can be found in this package. Making the best of both the MM algorithm and the Newton Raphson method, we derive stable and efficient algorithms to find the maximum likelihood estimates. If running on a multi-core machine, multi-threading operations are enabled.

Donwload R package

Download Matlab Toolbox

correlogram

In movies, whenever a criminal’s finger print is found, he is done. The policeman will magically find this person in their database in an eye blink. I was sad to learn that this is no more than a fantasy created by the directors. It is much harder and time consuming to match the finger print found in the criminal scene with those in the database. This particular R package is used to match finger print data utilizing the grascale and the pixel location information.

The package can also be used for other image processing problems, such as image categorization.