资源简介
模式识别-特征选择-几种算法的matlab实现。

代码片段和文件信息
%==================================================================
%--- Backward Feature selection by Correst Classification Rate ----
%--------------- Main Function ---------------------------------
%==================================================================
% function [ResultMat] = BackSel_main
% Matlab function of the two backward selection algorithms:
% a) Sequential Backward Selection (SBS)
% b) Sequential Floating Backward Selection (SFBS)
% Methods are improved by a t-test and a Information Loss
% evaluation. The criterion in feature selection is the correct
% classification achieved by the Bayes classifier when each
% probability density function is modeled as single Gaussian.
% Main function:
% - The feature selection method and our improvement on sequential
% selection algorithms
% Secondary function: BayesClassMVGaussPDFs
% - Bayes Classifier with Gaussian modeled PDFs
% using Crossvalidation or Resubstitution methods.
% Copyright 2003-2009 Dimitrios Ververidis AIIA Lab.
% $Revision: 5.1.4 $ $Date: 27/01/2009 $
% REFERENCES:
% [1] D. Ververidis and C. Kotropoulos “Fast and accurate feature
% subset selection applied into speech emotion recognition“
% Els. Signal Process. vol. 88 issue 12 pp. 2956-2970 2008.
% [2] D. Ververidis and C. Kotropoulos “Optimum feature subset
% selection with respect to the information loss of the
% Mahalanobis distance in high dimensions“ under preparation
% 2009.
function [ResultMat ConfMatFinal Tlapse OptimumFeatureSet...
OptimumError] = BackSel_main(DatasetToUse FSSettings handles)
format short g
% INPUT
% DatasetToUse: STRING (‘finalvecXXX‘ where XXX=Your dbname)
% FSSettings;
ErrorEstMethod = FSSettings.ErrorEstMethod;
FSMethod = FSSettings.FSMethod;
MahalInfoLossMethod = FSSettings.MahalInfoLossMethod;
GammaParam = FSSettings.GammaParam;
PercTest = FSSettings.PercTest;
ConfMatSwitch = FSSettings.ConfMatSwitch;
TotalNStepsThres = FSSettings.TotalNStepsThres;
LogViewOfIntStep = FSSettings.LogViewOfIntStep;
% ErrorEstMethod: Error Estimation Method (STRING)
% Values: ‘Standard‘ stands for cross-validation
% ‘ProposedA‘ stands for cross-validation
% ‘ProposedAB‘ stands for cross-validation
% (see paper [1])
% ‘Resubstitution‘ Train and Test sets are
% the whole data set
% MahalInfoLossMethod: (Binary) To use Limits of CCR wrt
% Dimensionality found with Mahalanobis Info
% Loss (see [2]) (DEFAULT = 1)
% FSMethod : (STRING) Feature Selection method (‘SFFS‘‘SFS‘)
% (DEFAULT = ‘SFS‘)
% PercTest: (Integer) 1/PercTest of data is used for testing.
% Remain is used for training the
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
文件 1369 2010-08-29 04:04 license.txt
目录 0 2014-03-12 14:05 Version_5.1.8_Out\
文件 23988 2009-03-07 14:01 Version_5.1.8_Out\BackSel_main.m
文件 9497 2009-02-01 19:51 Version_5.1.8_Out\BayesClassMVGaussPDFs.m
文件 2371 2009-01-17 17:38 Version_5.1.8_Out\BayesClassValidationSet.m
文件 4422 2009-02-20 12:45 Version_5.1.8_Out\CalcInfoLoss.m
文件 7915 2009-01-09 12:35 Version_5.1.8_Out\CCRForOptSet.m
文件 5432 2009-02-26 19:44 Version_5.1.8_Out\DataLoadAndPreprocess.m
文件 19208 2009-03-07 14:11 Version_5.1.8_Out\DEMO.fig
文件 12433 2009-03-07 14:56 Version_5.1.8_Out\DEMO.m
文件 20375 2009-03-07 14:01 Version_5.1.8_Out\ForwSel_main.m
目录 0 2014-03-12 14:05 Version_5.1.8_Out\Help\
文件 69120 2009-03-05 13:29 Version_5.1.8_Out\Help\AuthorsVerveridis.doc
文件 33198 2009-03-05 14:07 Version_5.1.8_Out\Help\AuthorsVerveridis.htm
文件 2487 2006-04-18 13:32 Version_5.1.8_Out\Help\dimsmall.jpg
文件 522240 2009-03-05 13:25 Version_5.1.8_Out\Help\Readme.doc
目录 0 2014-03-12 14:05 Version_5.1.8_Out\Help\Readme.files\
文件 196 2009-03-05 13:26 Version_5.1.8_Out\Help\Readme.files\filelist.xm
文件 488380 2009-03-05 13:26 Version_5.1.8_Out\Help\Readme.files\image001.jpg
文件 53345 2009-03-05 13:26 Version_5.1.8_Out\Help\Readme.files\image002.jpg
文件 18676 2009-03-05 13:26 Version_5.1.8_Out\Help\Readme.htm
文件 511517 2009-02-24 00:24 Version_5.1.8_Out\Help\Verver_ElsSigPro_08.pdf
目录 0 2014-03-12 14:05 Version_5.1.8_Out\icons\
文件 9544 2008-02-04 15:34 Version_5.1.8_Out\icons\PlayBlueIcon.tif
文件 9596 2008-02-04 15:36 Version_5.1.8_Out\icons\StopBlueIcon.tif
文件 8192 2009-01-14 19:13 Version_5.1.8_Out\icons\Thumbs.db
文件 2938 2009-02-04 16:31 Version_5.1.8_Out\LoadFromTXTfiles.m
目录 0 2014-03-12 14:05 Version_5.1.8_Out\LowLevelFunctions\
文件 1105 2008-03-24 16:38 Version_5.1.8_Out\LowLevelFunctions\FactorGamma.m
文件 1391 2009-01-12 15:51 Version_5.1.8_Out\LowLevelFunctions\ShowConfMat.m
目录 0 2014-03-12 14:05 Version_5.1.8_Out\PatTargMatrices\
............此处省略7个文件信息
相关资源
- Matlab路面裂缝识别69319
- 高灵敏度GPS接收机MATLAB仿真,附捕获
- 基于MATLAB的质点弹道计算与外弹道优
- 阵列天线的matlab仿真
- MATLAB 经典程序源代码大全
- MATLAB小波软阈值去噪代码33473
- 天线阵的波束形成在MATLAB仿真程序及
- 非线性SVM算法-matlab实现
- 《MATLAB 智能算法超级学习手册》-程序
- 组合导航matlab程序
- 读取txt文件内容matlab代码实现
- Matlab实现基于相关的模板匹配程序
- matlab优化工具箱讲解
- 基于MATLAB的快速傅里叶变换
- 光纤传输中的分布傅立叶算法matlab实
- 基于matlab的图像处理源程序
- matlab 椭圆拟合程序
- 算术编码解码matlab源代码
- optical_flow 光流法 matlab 实现程序
- 引导图像滤波器 Matlab实现
- 分形几何中一些经典图形的Matlab画法
- OFDM系统MATLAB仿真代码
- SVM工具箱(matlab中运行)
- 图像小波变换MatLab源代码
- LU分解的MATLAB实现
- 冈萨雷斯数字图像处理matlab版(第三
- 替代数据法的matlab程序
- 用matlab实现的多站定位系统性能仿真
- 通过不同方法进行粗糙集属性约简m
- k近邻算法matlab实现
评论
共有 条评论