基于机器学习算法的特征值分类附MATLAB代码R2019b.zip

function [D]=EigenClass(Train_Data,Training_Class,Test_Data,k) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % Code for paper: Erkan, U., A Precise and Stable Machine Learning Algorithm: % % Eigenvalue Classification (EigenClass). Neural Computing and Applications,Accept % % Volume, issue and page number will be added later. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Train_Data = training data % Training_Class = Classes of training data % Test_Data = Test data, excluding classes % k=the number of eigen values to be used Class_Unique=unique(Training_Class); Train_Data(Train_Data == 0) = 0.0001; Test_Data(Test_Data == 0) =0.0001; Samples_Number=size(Train_Data,2); for r=1:numel(Class_Unique) b=Train_Data((Training_Class == Class_Unique(r)),:); for i=1:size(Test_Data,1) for j=1:size(b,1) B(i,j,r)=sum(abs(1-eig(diag(b(j,1:Samples_Number)),diag(Test_Data(i,1:Samples_Number))))); end end B(:,:,r)= sort(B(:,:,r),2); C(:,r)=mean(B(:,1:k,r),2); clear B; end for i=1:size(Test_Data,1) [~,I] =min(C(i,:)); D(i,1)=Class_Unique(I); end end