Package: sg
Version: 0.4.4
Date: 2007-11-23
Title: The SHOGUN Machine Learning Toolbox
Author: Soeren Sonnenburg, Gunnar Raetsch, Fabio De Bona
Maintainer: Soeren Sonnenburg <Soeren.Sonnenburg@first.fraunhofer.de>
Depends: R (>= 2.1.0)
Suggests:
Description: The SHOGUN machine learning toolbox's focus is on kernel methods and especially on Support Vector Machines (SVM). It provides a generic SVM object interfacing to several different SVM implementations, among them the state of the art LibSVM[1] and SVMlight[2].  Each of the SVMs can be combined with a variety of kernels. The toolbox not only provides efficient implementations of the most common kernels, like the Linear, Polynomial, Gaussian and Sigmoid Kernel but also comes with a number of recent string kernels as e.g. the Locality Improved[3], Fischer[4], TOP[5], Spectrum[6], Weighted Degree Kernel (with shifts)[7]. For the latter the efficient LINADD[8] optimizations are implemented.  Also SHOGUN offers the freedom of working with custom pre-computed kernels.  One of its key features is the ``combined kernel'' which can be constructed by a weighted linear combination of a number of sub-kernels, each of which not necessarily working on the same domain. An optimal sub-kernel weighting can be learned using Multiple Kernel Learning[9].  Currently SVM 2-class classification and regression problems can be dealt with. However SHOGUN also implements a number of linear methods like Linear Discriminant Analysis (LDA), Linear Programming Machine (LPM), (Kernel) Perceptrons and features algorithms to train hidden markov models.  The input feature-objects can be dense, sparse or strings and of type int/short/double/char and can be converted into different feature types.  Chains of ``preprocessors'' (e.g. substracting the mean) can be attached to each feature object allowing for on-the-fly pre-processing.
License: GPL Version 2 or later.
URL: http://www.r-project.org, http://www.fml.tuebingen.mpg.de/raetsch/projects/shogun
