多属性决策分析案例结果.docVIP

多属性决策分析案例 (分析过程文件见文件夹“分析过程”) 第一步：利用MATLAB求综合属性值M-FILE源文件见“multiattribute.m” 代码： %first import the data for A!!! %首先输入数组 A（即文件“多属性决策”）! 源数据顺序经过了调整，把逆指标放在前23列，正指标放在后八列 R=zeros(60,31); R1=zeros(60,31); E=zeros(1,31); W=zeros(1,31); Z=zeros(60,1); B=ones(1,31);%把B变成全为1的数组 k=1:23; B(1,k)=1./[max(A(:,k))-min(A(:,k))]; %确定MAX(a(i,j))-MIN(a(i,j))数组 for j=1:23 R(:,j)=(max(A(:,j))-A(:,j)).*B(1,j);%先把前23个逆指标规范化 end B=ones(1,31);%再次把B变成全为1的数组 k=24:31; B(1,k)=1./[max(A(:,k))-min(A(:,k))]; for j=24:31 R(:,j)=(A(:,j)-min(A(:,j))).*B(1,j)%再把后八个正指标规范化 for j=1:31 R1(:,j)=R(:,j)./(sum(R(:,j)));%将矩阵R列归一化 end e=0; for j=1:1:31 for i=1:1:60 if R1(i,j)==0 s=0; else s=R1(i,j)*log(R1(i,j)); end e=s+e; end E(j)=(-1/log(60))*e;%得出信息熵 end sK=sum(1-E(1,:)); for j=1:31 W(j)=(1-E(j))./sK;%得出权重 end for i=1:60;z=0; for j=1:31 z=z+R(i,j)*W(j); end Z(i)=z;%得出综合属性值Z end end 综合属性值Z的具体数值见文件EXCLE “综合属性Z的数值“ 第二步：用SPSS进行聚类分析 分析结果 聚类成员 案例号 VAR00002 聚类 距离 1 1.00 1 .013 2 2.00 5 .002 3 3.00 4 .014 4 4.00 4 .001 5 5.00 4 .031 6 6.00 4 .018 7 7.00 5 .024 8 8.00 5 .017 9 9.00 5 .022 10 10.00 4 .012 11 11.00 2 .018 12 12.00 3 .006 13 13.00 5 .016 14 14.00 4 .000 15 15.00 4 .014 16 16.00 5 7.511E-6 17 17.00 2 .001 18 18.00 2 .010 19 19.00 2 .022 20 20.00 1 .019 21 21.00 4 .037 22 22.00 3 .002 23 23.00 1 .006 24 24.00 4 .036 25 25.00 2 .002 26 26.00 4 .016 27 27.00 4 .019 28 28.00 3 .001 29 29.00 4 .010 30 30.00 2 .030 31 31.00 2 .014 32 32.00 5 .002 33 33.00 3 .009 34 34.00 4 .024 35 35.00 4 .023 36 36.00 2 .017 37 37.00 5 .010 38 38.00 4 .002 39 39.00 2 .016 40 40.00 5 .007 41 41.00 2 .002 42 42.00 4 .005 43 43.00 2 .007 44 44.00 4 .003 45 45.00 2 .032 46 46.00 4 .021 47 47.00 4 .024 48 48.00 2 .023 49 49.00 2 .