Introduction
This page contains download of an efficient matlab toolbox containing useful statistical learning tools. It is strongly object oriented, and well documented. This toolbox is designed to allow an easy manipulation of probability models. It helps to build new statitical methods with very clean code.
Main features
Unsupervised Learning:
Multidimensional distributions,
parametric density models (or generative models),
Mixture distributions,
Learning/estimating the parameters:
Standard Maximum Likelihood estimators,
EM algorithm for latent class models,
Conjugate Gradient descent other probabilty models,
Model Selection (BIC criterion).
Supervised Learning:
Support Vector Machines (in conjunction with OSU-SVM toolbox),
K Nearest Neighbors (KNN) classification method,
Generative classifiers (also called Bayesian Classifiers), including Linear and Quadratic Discriminant Analysis, Mixture Discriminant Analysis
Model Selection, using Cross-Validation or other criterions.