模糊數(shù)學(xué)聚類分析-matlab作聚類分析圖

實(shí)驗(yàn)一：聚類分析要求：把照片分類一例（即例3），編程序做一遍。

照片分類?，F(xiàn)有三個(gè)家庭，每個(gè)家庭由四至七人組成，每人一張照片，共有十六張。通過(guò)照片按相貌相像程度分類，把三個(gè)家庭區(qū)分開來(lái)。首先，建立相似關(guān)系。任取兩張照片，請(qǐng)若干中學(xué)生按相貌相像程度打分，取平均數(shù)再折合成隸屬度，得到相像關(guān)系的F矩陣R（見表）由于矩陣是對(duì)稱的，只寫出下三角。101001000.4100.80010.500.20.20100.8000.4010.40.20.20.500.80100.400.80.40.20.401000.20.2000.200.2100.50.20.2000.800.40.21000.20.800000.40.8010.800.20.400.400.40000100.800.20.400.800.20.20.6001000.40.800.2000.2000.20.2010.6000.20.20.800.400000.40.20.41表1模糊矩陣A=[100000.500.400000.8000.6;

01000.800.80.20.400.5000.800;

0010.400.200.200.20.20.20.200.40;

000.4100.200.50.80.20.20.80.40.20.80.2;

00.800100.400.400000.400.2;

0.500.20.20100.80.20000.400.20.8;

00.8000.40100.40.20.8000.800;

0.40.20.20.500.80100000.4000.4;

00.400.80.40.20.4010.20.40.400.20.20;

000.20.2000.200.210.20.800.200;

00.50.20.2000.800.40.21000.600;

000.20.800000.40.801000.20;

0.800.20.400.400.40000100.20.4;

00.800.20.400.800.20.20.600100.2;

000.40.800.2000.2000.20.2010.4;

0.6000.20.20.800.400000.40.20.41]模糊矩陣合成令B=A，等于題目中給出的模糊矩陣，做模糊矩陣的合成運(yùn)算。Matlab程序編寫如下：M-文件見Max-Min(隨文件夾)模糊矩陣合成運(yùn)算程序function[C]=Max_Min(A,B)A=[];B=[];if(s1~=s)return;end[m,s]=size(A);[s1,n]=size(B);C=[];for(i=1:m)for(j=1:n)C(i,j)=0;for(k=1:s)x=0;if(A(i,k)<B(k,j))x=A(i,k);elsex=B(k,j);endif(C(i,j)<x)C(i,j)=x;endendend;end建立M文件如下：ans=Columns1through101.00000.20000.20000.40000.20000.600000.50000.200000.20001.00000.20000.40000.80000.20000.80000.20000.40000.20000.20000.20001.00000.400000.20000.20000.40000.40000.20000.40000.40000.40001.00000.40000.50000.40000.50000.80000.80000.20000.800000.40001.00000.20000.80000.20000.40000.20000.60000.20000.20000.50000.20001.00000.20000.80000.20000.200000.80000.20000.40000.80000.20001.00000.20000.40000.20000.50000.20000.40000.50000.20000.80000.20001.00000.50000.20000.20000.40000.40000.80000.40000.20000.40000.50001.00000.400000.20000.20000.80000.20000.20000.20000.20000.40001.000000.80000.20000.40000.50000.20000.80000.20000.40000.200000.40000.40000.80000.40000.20000.40000.50000.80000.80000.80000.20000.40000.40000.20000.500000.40000.40000.20000.20000.80000.20000.20000.80000.20000.80000.20000.40000.20000.40000.20000.40000.80000.20000.40000.20000.50000.80000.20000.60000.20000.40000.40000.20000.80000.20000.80000.20000.2000Columns11through16000.80000.20000.40000.60000.80000.40000.20000.80000.20000.20000.20000.40000.40000.20000.40000.40000.40000.80000.40000.20000.80000.40000.50000.40000.20000.80000.20000.20000.20000.20000.50000.20000.40000.80000.80000.400000.80000.20000.20000.20000.50000.40000.20000.50000.80000.40000.80000.40000.40000.80000.20000.20000.80000.20000.20000.20000.20001.00000.40000.20000.80000.20000.20000.40001.00000.40000.20000.80000.20000.20000.40001.00000.20000.40000.60000.80000.20000.20001.00000.20000.20000.20000.80000.40000.20001.00000.40000.20000.20000.60000.20000.40001.0000在Matlab中輸入如下命令：>>Max_Min將第一次矩陣合成運(yùn)算的矩陣記為C對(duì)比：模糊矩陣C不等于A,故該矩陣不是傳遞閉包。進(jìn)行第二次合成運(yùn)算：第一次做模糊矩陣合成運(yùn)算的矩陣C,令D=C,編寫如下程序：（見M文件Max_Min2.m(隨文件夾)）運(yùn)行結(jié)果如下：E=1.00.40.40.50.40.60.40.60.50.40.40.50.80.20.50.60.41.00.40.40.80.40.80.40.40.40.80.40.40.80.40.40.40.41.00.40.40.40.40.40.40.40.40.40.40.40.40.40.50.40.41.00.40.50.40.50.80.80.40.80.50.40.80.50.40.80.40.41.00.40.80.40.40.40.80.40.40.80.40.40.60.40.40.50.41.00.40.80.50.50.40.50.60.20.50.80.40.80.40.40.80.41.00.40.40.40.80.40.40.80.40.40.60.40.40.50.40.80.41.00.50.50.40.50.60.40.50.80.50.40.40.80.40.50.40.51.00.80.40.80.40.40.80.50.40.40.40.80.40.50.40.50.81.00.40.80.40.40.80.40.40.80.40.40.80.40.80.40.40.41.00.40.40.80.40.40.50.40.40.80.40.50.40.50.80.80.41.00.40.40.80.50.80.40.40.50.40.60.40.60.40.40.40.41.00.40.40.60.20.80.40.40.80.20.80.40.40.40.80.40.41.00.40.20.50.40.40.80.40.50.40.50.80.80.40.80.40.41.00.50.60.40.40.50.40.80.40.80.50.40.40.50.60.20.51.0經(jīng)對(duì)比,模糊矩陣E不等于矩陣C，不是傳遞閉包。進(jìn)行第三次矩陣合成運(yùn)算，具體程序見M文件Max_Min3.m(隨文件夾)經(jīng)Matlab實(shí)現(xiàn)如下：>>Max_Min3ans=Columns1through81.00000.40000.40000.50000.40000.60000.40000.60000.40001.00000.40000.40000.80000.40000.80000.40000.40000.40001.00000.40000.40000.40000.40000.40000.50000.40000.40001.00000.40000.50000.40000.50000.40000.80000.40000.40001.00000.40000.80000.40000.60000.40000.40000.50000.40001.00000.40000.80000.40000.80000.40000.40000.80000.40001.00000.40000.60000.40000.40000.50000.40000.80000.40001.00000.50000.40000.40000.80000.40000.50000.40000.50000.50000.40000.40000.80000.40000.50000.40000.50000.40000.80000.40000.40000.80000.40000.80000.40000.50000.40000.40000.80000.40000.50000.40000.50000.80000.40000.40000.50000.40000.60000.40000.60000.40000.80000.40000.40000.80000.40000.80000.40000.50000.40000.40000.80000.40000.50000.40000.50000.60000.40000.40000.50000.40000.80000.40000.8000Columns9through160.50000.50000.40000.50000.80000.40000.50000.60000.40000.40000.80000.40000.40000.80000.40000.40000.40000.40000.40000.40000.40000.40000.40000.40000.80000.80000.40000.80000.50000.40000.80000.50000.40000.40000.80000.40000.40000.80000.40000.40000.50000.50000.40000.50000.60000.40000.50000.80000.40000.40000.80000.40000.40000.80000.40000.40000.50000.50000.40000.50000.60000.40000.50000.80001.00000.80000.40000.80000.50000.40000.80000.50000.80001.00000.40000.80000.50000.40000.80000.50000.40000.40001.00000.40000.40000.80000.40000.40000.80000.80000.40001.00000.50000.40000.80000.50000.50000.50000.40000.50001.00000.40000.50000.60000.40000.40000.80000.40000.40001.00000.40000.40000.80000.80000.40000.80000.50000.40001.00000.50000.50000.50000.40000.50000.60000.40000.50001.0000記此矩陣為F,F不等于E,故F不是傳遞閉包，進(jìn)行第四次矩陣合成運(yùn)算：具體程序見M文件Max_Min4.m(隨文件夾)Matlab實(shí)現(xiàn)結(jié)果如下：記第四次矩陣合成運(yùn)算的結(jié)果為G>>Max_Min4ans=Columns1through81.00000.40000.40000.50000.40000.60000.40000.60000.40001.00000.40000.40000.80000.40000.80000.40000.40000.40001.00000.40000.40000.40000.40000.40000.50000.40000.40001.00000.40000.50000.40000.50000.40000.80000.40000.40001.00000.40000.80000.40000.60000.40000.40000.50000.40001.00000.40000.80000.40000.80000.40000.40000.80000.40001.00000.40000.60000.40000.40000.50000.40000.80000.40001.00000.50000.40000.40000.80000.40000.50000.40000.50000.50000.40000.40000.80000.40000.50000.40000.50000.40000.80000.40000.40000.80000.40000.80000.40000.50000.40000.40000.80000.40000.50000.40000.50000.80000.40000.40000.50000.40000.60000.40000.60000.40000.80000.40000.40000.80000.40000.80000.40000.50000.40000.40000.80000.40000.50000.40000.50000.60000.40000.40000.50000.40000.80000.40000.8000Columns9through160.50000.50000.40000.50000.80000.40000.50000.60000.40000.40000.80000.40000.40000.80000.40000.40000.40000.40000.40000.40000.40000.40000.40000.40000.80000.80000.40000.80000.50000.40000.80000.50000.40000.40000.80000.40000.40000.80000.40000.40000.50000.50000.40000.50000.60000.40000.50000.80000.40000.40000.80000.40000.40000.80000.40000.40000.50000.50000.40000.50000.60000.40000.50000.80001.00000.80000.40000.80000.50000.40000.80000.50000.80001.00000.40000.80000.50000.40000.80000.50000.40000.40001.00000.40000.40000.80000.40000.40000.80000.80000.40001.00000.50000.40000.80000.50000.50000.50000.40000.50001.00000.40000.50000.60000.40000.40000.80000.40000.40001.00000.40000.40000.80000.80000.40000.80000.50000.40001.00000.50000.50000.50000.40000.50000.60000.40000.50001.0000矩陣合成運(yùn)算求得傳遞閉包對(duì)比矩陣F,可知G=F,則矩陣F為模糊矩陣A的傳遞閉包。傳遞閉包由以上分析的模糊矩陣A的傳遞閉包為t(A)=[1.00.40.40.50.40.60.40.60.50.50.40.50.80.40.50.60.41.00.40.40.80.40.80.40.40.40.80.40.40.80.40.40.40.41.00.40.40.40.40.40.40.40.40.40.40.40.40.40.50.40.41.00.40.50.40.50.80.80.40.80.50.40.80.50.40.80.40.41.00.40.80.40.40.40.80.40.40.80.40.40.60.40.40.50.41.00.40.80.50.50.40.50.60.40.50.80.40.80.40.40.80.41.00.40.40.40.80.40.40.80.40.40.60.40.40.50.40.80.41.00.50.50.40.50.60.40.50.80.50.40.40.80.40.50.40.51.00.80.40.80.50.40.80.50.50.40.40.80.40.50.40.50.81.00.40.80.50.40.80.50.40.80.40.40.80.40.80.40.40.41.00.40.40.80.40.40.50.40.40.80.40.50.40.50.80.80.41.00.50.400.80.50.80.40.40.50.40.60.40.60.50.50.40.51.00.40.50.60.40.80.40.40.80.40.80.40.40.40.80.40.41.00.40.40.50.40.40.80.40.50.40.50.80.80.40.80.50.41.00.50.60.40.40.50.40.80.40.80.50.50.40.50.60.40.51.0]聚類分析建立M文件，如下M文件juleifenxi.m(隨文件夾)（這里分類取為4）gj=[1.00.40.40.50.40.60.40.60.50.50.40.50.80.40.50.60.41.00.40.40.80.40.80.40.40.40.80.40.40.8