Code covered by the BSD License  

Highlights from
Multi Class Support Vector Machine

image thumbnail

Multi Class Support Vector Machine

by

 

07 Oct 2011 (Updated )

This function removes out the limitation of MATLAB SVM function of two class and uses more classes.

multisvm( T,C,test )
function [itrfin] = multisvm( T,C,test )
%MULTISVM(3.0) classifies the class of given training vector according to the 
% given group and gives us result that which class it belongs.
% We have also to input the testing matrix

%Inputs: T=Training Matrix, C=Group, test=Testing matrix
%Outputs: itrfin=Resultant class(Group,USE ROW VECTOR MATRIX) to which tst set belongs 

%----------------------------------------------------------------------%
% IMPORTANT: DON'T USE THIS PROGRAM FOR CLASS LESS THAN 3,             %
%            OTHERWISE USE svmtrain,svmclassify DIRECTLY or            %
%            add an else condition also for that case in this program. %
%            Modify required data to use Kernel Functions and Plot also%
%----------------------------------------------------------------------%
%                       Date:11-08-2011(DD-MM-YYYY)                    %
% This function for multiclass Support Vector Machine is written by
% ANAND MISHRA (Machine Vision Lab. CEERI, Pilani, India) 
% and this is free to use. email: anand.mishra2k88@gmail.com

% Updated version 2.0 Date:14-10-2011(DD-MM-YYYY)
% Updated version 3.0 Date:04-04-2012(DD-MM-YYYY)

itrind=size(test,1);
itrfin=[];
Cb=C;
Tb=T;
for tempind=1:itrind
    tst=test(tempind,:);
    C=Cb;
    T=Tb;
    u=unique(C);
    N=length(u);
    c4=[];
    c3=[];
    j=1;
    k=1;
    if(N>2)
        itr=1;
        classes=0;
        cond=max(C)-min(C);
        while((classes~=1)&&(itr<=length(u))&& size(C,2)>1 && cond>0)
        %This while loop is the multiclass SVM Trick
            c1=(C==u(itr));
            newClass=c1;
            svmStruct = svmtrain(T,newClass,'kernel_function','rbf');   % I am using rbf kernel function, you must change it also
            classes = svmclassify(svmStruct,tst);
        
            % This is the loop for Reduction of Training Set
            for i=1:size(newClass,2)
                if newClass(1,i)==0;
                    c3(k,:)=T(i,:);
                    k=k+1;
                end
            end
        T=c3;
        c3=[];
        k=1;
        
            % This is the loop for reduction of group
            for i=1:size(newClass,2)
                if newClass(1,i)==0;
                    c4(1,j)=C(1,i);
                    j=j+1;
                end
            end
        C=c4;
        c4=[];
        j=1;
        
        cond=max(C)-min(C); % Condition for avoiding group 
                            %to contain similar type of values 
                            %and the reduce them to process
        
            % This condition can select the particular value of iteration
            % base on classes
            if classes~=1
                itr=itr+1;
            end    
        end
    end

valt=Cb==u(itr);		% This logic is used to allow classification
val=Cb(valt==1);		% of multiple rows testing matrix
val=unique(val);
itrfin(tempind,:)=val;  
end

end

% Give more suggestions for improving the program.

Contact us