资源简介
常见的一种特征筛选手段,可以从大量变量中筛选特征变量实现保留变量与目标之间的最大相关性,而彼此间的重复信息最小
代码片段和文件信息
function [fea] = mrmr_mid_d(d f K)
% function [fea] = mrmr_mid_d(d f K)
%
% The MID scheme of minimum redundancy maximal relevance (mRMR) feature selection
%
% The parameters:
% d - a N*M matrix indicating N samples each having M dimensions. Must be integers.
% f - a N*1 matrix (vector) indicating the class/category of the N samples. Must be categorical.
% K - the number of features need to be selected
%
% Note: This version only supports discretized data thus if you have continuous data in “d“ you
% will need to discretize them first. This function needs the mutual information computation
% toolbox written by the same author downloadable at the Matlab source codes exchange site.
% Also There are multiple newer versions on the Hanchuan Peng‘s web site
% (http://research.janelia.org/peng/proj/mRMR/index.htm).
%
% More information can be found in the following papers.
%
% H. Peng F. Long and C. Ding
% “Feature selection based on mutual information: criteria
% of max-dependency max-relevance and min-redundancy“
% IEEE Transactions on Pattern Analysis and Machine Intelligence
% Vol. 27 No. 8 pp.1226-1238 2005.
%
% C. Ding and H. Peng
% “Minimum redundancy feature selection from microarray gene
% expression data“
% Journal of Bioinformatics and Computational Biology
% Vol. 3 No. 2 pp.185-205 2005.
%
% C. Ding and
评论
共有 条评论