Feature Selection Package - Algorithms- BLogReg
Description
This algorithm is an implementation of a Bayesian logistic regression method based on Shevade & Keerthi's algorithm. blogreg.c is the MEX implementaiton written by Dr. Gavin C. Cawley (paper below), and fsblogreg.m is the MATLAB wrapper for that code.
Usage
Method Signature:
[out] = fsblogreg(X, Y, param)

Output:
   out: The selected features of the algorithm, represented as an array of numbers.

Input:
   X: The list of data points, each row being an instance.
   Y: The list of data points, each column being a class.
   param: A struct with the field 'tol', that contains the value you wish to use for the tolerance.
Code Example
% Using the BASEHOCK.m test data set that is
% included in the package.
param.tol = 1;
fsblogreg(X,Y,param)
Keyword in Evaluator Framework
blogreg
Paper
BibTex entry for:

Gene selection in cancer classification using sparse logistic regression with Bayesian regularisation
@article{Cawley06,
   author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
   title = {Gene selection in cancer classification using sparse logistic regression with Bayesian regularization},
   journal = {Bioinformatics},
   volume={22},
   number={19},
   year={2006},
   issn={1367-4803},
   pages={2348--2355}
}