How to realize intrusion clustering of GRNN Network based on Generalized Neural Network by matlab
This article mainly introduces matlab how to achieve GRNN network intrusion clustering based on generalized neural network related knowledge, detailed and easy to understand, simple and fast operation, with certain reference value, I believe that everyone reading this matlab how to achieve GRNN network intrusion clustering based on generalized neural network articles will have some gains, let's take a look at it.
%% Empty environment files
clear
clc
%% Extract attack data
% Attack Sample Data
load netattack;
P1=netattack;
T1=P1(:,39)';
P1(:,39)=[];
% Data Size
[R1,C1]=size(P1);
csum=20; % How much training data is extracted
%% Fuzzy Clustering
data=P1;
[center,U,obj_fcn] = fcm(data,5);
for i=1:R1
[value,idx]=max(U(:,i));
a1(i)=idx;
end
%% Fuzzy Cluster Result Analysis
Confusion_Matrix_FCM=zeros(6,6);
Confusion_Matrix_FCM(1,:)=0:5;
Confusion_Matrix_FCM(:,1)=[0:5]';
for nf=1:5
for nc=1:5
Confusion_Matrix_FCM(nf+1,nc+1)=length(find(a1(T1==nf)==nc));
end
end
%% Network training sample extraction
cent1=P1(a1==1,:);cent1=mean(cent1);
cent2=P1(a1==2,:);cent2=mean(cent2);
cent3=P1(a1==3,:);cent3=mean(cent3);
cent4=P1(a1==4,:);cent4=mean(cent4);
cent5=P1(a1==5,:);cent5=mean(cent5);
% Extract norm minimum for training samples
for n=1:R1
ecent1(n)=norm(P1(n,:)-cent1);
ecent2(n)=norm(P1(n,:)-cent2);
ecent3(n)=norm(P1(n,:)-cent3);
ecent4(n)=norm(P1(n,:)-cent4);
ecent5(n)=norm(P1(n,:)-cent5);
end
for n=1:csum
[~, me1]=min(ecent1);
[~, me2]=min(ecent2);
[~, me3]=min(ecent3);
[~, me4]=min(ecent4);
[va, me5]=min(ecent5);
ecnt1(n,:)=P1(me1(1),:);ecent1(me1(1))=[];tcl(n)=1;
ecnt2(n,:)=P1(me2(1),:);ecent2(me2(1))=[];tc2(n)=2;
ecnt3(n,:)=P1(me3(1),:);ecent3(me3(1))=[];tc3(n)=3;
ecnt4(n,:)=P1(me4(1),:);ecent4(me4(1))=[];tc4(n)=4;
ecnt5(n,:)=P1(me5(1),:);ecent5(me5(1))=[];tc5(n)=5;
end
P2=[ecnt1;ecnt2;ecnt3;ecnt4;ecnt5];T2=[tcl,tc2,tc3,tc4,tc5];
k=0;
%% iterative calculation
for nit=1:10% Start iteration
%% Generalized Neural Network Clustering
net = newgrnn(P2',T2,50);
% Training Generalized Networks
a2=sim(net,P1') ;
% Forecast Results
% Output Normalized (Classified by Output)
a2(a21.5&a22.5&a23.5&a24.5)=5;
%% Network training data extracted again
cent1=P1(a2==1,:);cent1=mean(cent1);
cent2=P1(a2==2,:);cent2=mean(cent2);
cent3=P1(a2==3,:);cent3=mean(cent3);
cent4=P1(a2==4,:);cent4=mean(cent4);
cent5=P1(a2==5,:);cent5=mean(cent5);
for n=1:R1% Calculate the distance from the sample to each center
ecent1(n)=norm(P1(n,:)-cent1);
ecent2(n)=norm(P1(n,:)-cent2);
ecent3(n)=norm(P1(n,:)-cent3);
ecent4(n)=norm(P1(n,:)-cent4);
ecent5(n)=norm(P1(n,:)-cent5);
end
% Select csum samples closest to the center of each class
for n=1:csum
[~, me1]=min(ecent1);
[~, me2]=min(ecent2);
[~, me3]=min(ecent3);
[~, me4]=min(ecent4);
[va, me5]=min(ecent5);
ecnt1(n,:)=P1(me1(1),:);ecent1(me1(1))=[];tc1(n)=1;
ecnt2(n,:)=P1(me2(1),:);ecent2(me2(1))=[];tc2(n)=2;
ecnt3(n,:)=P1(me3(1),:);ecent3(me3(1))=[];tc3(n)=3;
ecnt4(n,:)=P1(me4(1),:);ecent4(me4(1))=[];tc4(n)=4;
ecnt5(n,:)=P1(me5(1),:);ecent5(me5(1))=[];tc5(n)=5;
end
p2=[ecnt1;ecnt2;ecnt3;ecnt4;ecnt5];
T2=[tc1,tc2,tc3,tc4,tc5];
% Statistical classification results
Confusion_Matrix_GRNN=zeros(6,6);
Confusion_Matrix_GRNN(1,:)=0:5;
Confusion_Matrix_GRNN(:,1)=[0:5]';
for nf=1:5
for nc=1:5
Confusion_Matrix_GRNN(nf+1,nc+1)=length(find(a2(T1==nf)==nc));
end
end
pre2=0;
for n=2:6
pre2=pre2+max(Confusion_Matrix_GRNN(n,:));
end
pre2=pre2/R1*100;
end
About "matlab how to achieve network intrusion clustering based on generalized neural network GRNN" The content of this article is introduced here, thank you for reading! I believe everyone has a certain understanding of "matlab how to achieve network intrusion clustering based on generalized neural network GRNN" knowledge. If you still want to learn more knowledge, please pay attention to the industry information channel.
