%% 基于随机森林思想的组合分类器设计

%% 清空环境变量
close all;
clear;
clc;

%% 导入数据
Data=load(\'E:\study\研究生\实验\dataset\new_housing.txt\');
Label=load(\'E:\study\研究生\实验\dataset\new_housingLabel.txt\');
sum_Acc=0;
sum_MCC=0;
sum_F_measure=0;
sum_G_mean=0;
sum_AUC=0;
[M,N]=size(Data);%数据集为一个M*N的矩阵，其中每一行代表一个样本
indices=crossvalind(\'Kfold\',M,5);%进行随机分包
for k=1:5 %交叉验证k=5，5个包轮流作为测试集
    test = (indices == k); %获得test集元素在数据集中对应的单元编号
    train = ~test;%train集元素的编号为非test元素的编号
    train_data=Data(train,:);%从数据集中划分出train样本的数据
    train_target=Label(train,:);%获得样本集的测试目标，在本例中是实际分类情况
    test_data=Data(test,:);%test样本集
    test_target=Label(test,:);
    %模型与预测结果
    %% 创建随机森林分类器
    model = classRF_train(train_data,train_target);
    %% 仿真测试
    [Predict_label,votes] = classRF_predict(test_data,model);
    %预测结果概率输出
    prob_estimates=votes/500;%500为决策树数目
    output=prob_estimates(:,2);%预测为正类的概率
     %调整阈值进行预测和混淆矩阵的计算
    T=0.5;
    max_MCC=0;%记录最大的MCC值
    evaluation=[0,0,0,0];%四个评估指标存放地，初始化为全为0
    for T=0.1:0.01:0.9%步长0.01
        TP=0;
        FN=0;
        FP=0;
        TN=0;
        [r,~]=size(test_target);
        for i=1:r%样本个数
            if test_target(i,1)==1&&prob_estimates(i,2)>=T%本为正类，大于等于T则预测为正类 %正类的预测概率在prob_estimates第2列
                TP=TP+1;
            elseif test_target(i,1)==1&&prob_estimates(i,2)<T%本为正类，小于T则预测为负类
                FN=FN+1;
            elseif test_target(i,1)==-1&&prob_estimates(i,2)>=T%本为负类，大于等于T则预测为正类
                FP=FP+1;
            else %即tsetLabel(i,1)==-1&&prob_estimates(i,1)<T%本为负类小于T则预测为负类
                TN=TN+1;
            end
        end
        TP
        FN
        FP
        TN
        %Sen=TP/(TP+FN);
        %Spe=TN/(TN+FP);
        MCC=(TP*TN-FP*FN)/sqrt((TP+FP)*(TP+FN)*(TN+FP)*(TN+FN))
        if MCC>max_MCC%选择MCC值最大的那一组评估指标值
            max_MCC=MCC;
            Precision=TP/(TP+FP);
            Recall=TP/(TP+FN);
            TPR=TP/(TP+FN);
            TNR=TN/(TN+FP);
            Acc=(TP+TN)/(TP+TN+FP+FN);
            F_measure=(2*Precision*Recall)/(Precision+Recall);
            G_mean=sqrt(TPR*TNR);
            evaluation(1,1)=max_MCC;
            evaluation(1,2)=Acc;
            evaluation(1,3)=F_measure;
            evaluation(1,4)= G_mean;
        end
    end
    auc=AUC(test_target,output);%每一次分类结束后进行一次计算
    sum_AUC=sum_AUC+auc;
    sum_MCC=sum_MCC+ evaluation(1,1);
    sum_Acc=sum_Acc+ evaluation(1,2);
    sum_F_measure=sum_F_measure+ evaluation(1,3);
    sum_G_mean=sum_G_mean+evaluation(1,4);
end
avg_Acc=sum_Acc/5
avg_MCC= sum_MCC/5
avg_F_measure=sum_F_measure/5
avg_G_mean=sum_G_mean/5
avg_AUC=sum_AUC/5