matlab实现的ID3 分类决策树 算法

《matlab实现的ID3 分类决策树 算法》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。

1、functionD=ID3(train_features,train_targets,params,region)%ClassifyusingQuinlan'sID3algorithm%Inputs:%features-Trainfeatures%targets-Traintargets%params-[Numberofbinsforthedata,Percentageofincorrectlyassignedsamplesatanode]%region-Decisionregionvector:[-xx-yynumber_of_p

2、oints]%%Outputs%D-Decisionsufrace[Ni,M]=size(train_features);%·µ»ØÐÐÊýNiºÍÁÐÊýM%Getparameters[Nbins,inc_node]=process_params(params);inc_node=inc_node*M/100;%ForthedecisionregionN=region(5);mx=ones(N,1)*linspace(region(1),region(2),N);%linspace(Æðʼֵ£¬ÖÕÖ¹Öµ£¬ÔªËظöÊý

3、)my=linspace(region(3),region(4),N)'*ones(1,N);flatxy=[mx(:),my(:)]';%Preprocessing[f,t,UW,m]=PCA(train_features,train_targets,Ni,region);train_features=UW*(train_features-m*ones(1,M));flatxy=UW*(flatxy-m*ones(1,N^2));%First,binthedataandthedecisionregiondata[H,binned_

4、features]=high_histogram(train_features,Nbins,region);[H,binned_xy]=high_histogram(flatxy,Nbins,region);%Buildthetreerecursivelydisp('Buildingtree')tree=make_tree(binned_features,train_targets,inc_node,Nbins);%Makethedecisionregionaccordingtothetreedisp('Buildingdecisi

5、onsurfaceusingthetree')targets=use_tree(binned_xy,1:N^2,tree,Nbins,unique(train_targets));D=reshape(targets,N,N);%ENDfunctiontargets=use_tree(features,indices,tree,Nbins,Uc)%Classifyrecursivelyusingatreetargets=zeros(1,size(features,2));%size(features,2)·µ»ØfeaturesµÄÁ

6、ÐÊýif(size(features,1)==1),%Onlyonedimensionleft,soworkonitfori=1:Nbins,in=indices(find(features(indices)==i));if~isempty(in),ifisfinite(tree.child(i)),targets(in)=tree.child(i);else%Nodatawasfoundinthetrainingsetforthisbin,sochooseitrandomallyn=1+floor(rand(1)*length(

7、Uc));targets(in)=Uc(n);endendendbreakend%Thisisnotthelastlevelofthetree,so:%First,findthedimensionwearetoworkondim=tree.split_dim;dims=find(~ismember(1:size(features,1),dim));%Andclassifyaccordingtoitfori=1:Nbins,in=indices(find(features(dim,indices)==i));targets=targe

8、ts+use_tree(features(dims,:),in,tree.child(i),Nbins,Uc);end%ENDuse_treefunctiontree=make_tree(features,targets,inc_no