About: The opensource Cpackage fastICA implements the fastICA algorithm of Aapo Hyvarinen et al. (URL: http://www.cs.helsinki.fi/u/ahyvarin/) to perform Independent Component Analysis (ICA) and Projection Pursuit. fastICA is released under the GNU Public License (GPL). Changes:Initial Announcement on mloss.org.

About: Heteroscedastic Discriminant Analysis Changes:Fetched by rcranrobot on 20130401 00:00:05.551691

About: JProGraM (PRObabilistic GRAphical Models in Java) is a statistical machine learning library. It supports statistical modeling and data analysis along three main directions: (1) probabilistic graphical models (Bayesian networks, Markov random fields, dependency networks, hybrid random fields); (2) parametric, semiparametric, and nonparametric density estimation (Gaussian models, nonparanormal estimators, Parzen windows, NadarayaWatson estimator); (3) generative models for random networks (smallworld, scalefree, exponential random graphs, Fiedler random graphs/fields), subgraph sampling algorithms (random walk, snowball, etc.), and spectral decomposition. Changes:JProGraM 13.2  CHANGE LOG Release date: February 13, 2012 New features:  Support for Fiedler random graphs/random field models for largescale networks (ninofreno.graph.fiedler package);  Various bugfixes and enhancements (especially in the ninofreno.graph and ninofreno.math package).

About: Regularization paths for regression models with grouped covariates Changes:Fetched by rcranrobot on 20130401 00:00:05.489694

About: ModelBased Boosting Changes:Fetched by rcranrobot on 20130401 00:00:06.324985

About: L1 Regularization Path for Generalized Linear Models and Cox Proportional Hazards Model Changes:Fetched by rcranrobot on 20130401 00:00:05.305206

About: Generalized Boosted Regression Models Changes:Fetched by rcranrobot on 20130401 00:00:05.019963

About: A Laboratory for Recursive Partytioning Changes:Fetched by rcranrobot on 20130401 00:00:06.775432

About: Classification rule based on Bayesian naive Bayes models with feature selection bias corrected Changes:Fetched by rcranrobot on 20121201 00:00:07.510624

About: Bayesian Prediction with Highorder Interactions Changes:Fetched by rcranrobot on 20121201 00:00:03.777292

About: Mapping, pruning, and graphing tree models Changes:Fetched by rcranrobot on 20130401 00:00:06.263217

About: L1 constrained estimation aka `lasso' Changes:Fetched by rcranrobot on 20130401 00:00:05.967868

About: Improved Predictors Changes:Fetched by rcranrobot on 20130401 00:00:05.613011

About: Python Machine Learning Toolkit Changes:Added LASSO (using coordinate descent optimization). Made SVM classification (learning and applying) much faster: 2.5x speedup on yeast UCI dataset.

About: L1 (lasso and fused lasso) and L2 (ridge) penalized estimation in GLMs and in the Cox model Changes:Fetched by rcranrobot on 20130401 00:00:06.939105

About: Breiman and Cutler's random forests for classification and regression Changes:Fetched by rcranrobot on 20130401 00:00:07.638240

About: TurboParser is a free multilingual dependency parser based on linear programming developed by André Martins. It is based on joint work with Noah Smith, Mário Figueiredo, Eric Xing, Pedro Aguiar. Changes:This version introduces a number of new features:
Note: The runtimes above are approximate, and based on experiments with a desktop machine with a Intel Core i7 CPU 3.4 GHz and 8GB RAM. To run this software, you need a standard C++ compiler. This software has the following external dependencies: AD3, a library for approximate MAP inference; Eigen, a template library for linear algebra; googleglog, a library for logging; gflags, a library for commandline flag processing. All these libraries are free software and are provided as tarballs in this package. This software has been tested on Linux, but it should run in other platforms with minor adaptations.

About: Simpler use of data mining methods (e.g. NN and SVM) in classification and regression. Changes:Fetched by rcranrobot on 20130401 00:00:08.226306

About: MDP is a Python library of widely used data processing algorithms that can be combined according to a pipeline analogy to build more complex data processing software. The base of available algorithms includes signal processing methods (Principal Component Analysis, Independent Component Analysis, Slow Feature Analysis), manifold learning methods ([Hessian] Locally Linear Embedding), several classifiers, probabilistic methods (Factor Analysis, RBM), data preprocessing methods, and many others. Changes:What's new in version 3.3?
