Diff is too large to be shown
moepoi / Data Mining (UTS)
Last active 4 months ago
Diff is too large to be shown
| @@ -93,11 +93,12 @@ gbr28=reshape(gbr28',1,[]); | |||
| 93 | 93 | gbr29=reshape(gbr29',1,[]); | |
| 94 | 94 | gbr30=reshape(gbr30',1,[]); | |
| 95 | 95 | ||
| 96 | - | X = [gbr1, gbr2, gbr3, gbr4, gbr5, gbr6, gbr7, gbr8, gbr9, gbr10, gbr11, gbr12, gbr13, gbr14, gbr15, gbr16, gbr17, gbr18, gbr19, gbr20, gbr21, gbr22, gbr23, gbr24, gbr25, gbr26, gbr27, gbr28, gbr29, gbr30]; | |
| 96 | + | X = [gbr1; gbr2; gbr3; gbr4; gbr5; gbr6; gbr7; gbr8; gbr9; gbr10; gbr11; gbr12; gbr13; gbr14; gbr15; gbr16; gbr17; gbr18; gbr19; gbr20; gbr21; gbr22; gbr23; gbr24; gbr25; gbr26; gbr27; gbr28; gbr29; gbr30]; | |
| 97 | 97 | [n,p] = size(X); | |
| 98 | - | Xrata = mean(X) | |
| 98 | + | Xrata = mean(X, 1); | |
| 99 | 99 | ||
| 100 | 100 | %1. mengurangi setiap variable dengan rata2nya (normalisasi data) | |
| 101 | + | Z = zeros(size(X)); | |
| 101 | 102 | for i = 1:n | |
| 102 | 103 | Z(i, :) = X(i, :)-Xrata; | |
| 103 | 104 | end | |
| @@ -106,12 +107,13 @@ end | |||
| 106 | 107 | S = cov(X); % varian | |
| 107 | 108 | ||
| 108 | 109 | %Standardization Data | |
| 110 | + | Zr = zeros(size(Z)); | |
| 109 | 111 | for i = 1:n | |
| 110 | 112 | for j = 1:p | |
| 111 | 113 | Zr(i,j) = Z(i,j) / sqrt(S(j,j)); | |
| 112 | 114 | end | |
| 113 | 115 | end | |
| 114 | - | %ZR ADALAH HASIL DARI NORMALISASI DATA X (DATA AWAL) | |
| 116 | + | % Zr ADALAH HASIL DARI NORMALISASI DATA X (DATA AWAL) | |
| 115 | 117 | ||
| 116 | 118 | %2. Menghitung cov dari data yang sudah direduksi | |
| 117 | 119 | Zcov = cov(Zr); % matrix covarian dari variable hasil normalisasi Zr | |
| @@ -123,38 +125,40 @@ Zcov = cov(Zr); % matrix covarian dari variable hasil normalisasi Zr | |||
| 123 | 125 | % Hasil library matlab eig() --> hasil = diagonal utamanya urut dari kecil ke besar | |
| 124 | 126 | % yang diperlukan dari besar ke kecil | |
| 125 | 127 | %mengurutkan hasil eigen (DESC) | |
| 126 | - | m = 1 | |
| 128 | + | m = 1; | |
| 127 | 129 | for i = p:-1:1 | |
| 128 | 130 | DA_urut(m,m) = DA(i,i); | |
| 129 | - | VA_urut(m,:) = VA(i,:); | |
| 131 | + | VA_urut(m,:) = VA(:,i)'; | |
| 130 | 132 | m = m+1; | |
| 131 | 133 | end | |
| 132 | 134 | ||
| 133 | 135 | %menghitung komponen utama | |
| 136 | + | eValue = zeros(p, 1); | |
| 134 | 137 | for i = 1:p | |
| 135 | 138 | eValue(i,1) = DA_urut(i,i); | |
| 136 | 139 | end | |
| 137 | 140 | ||
| 138 | - | Total = sum(eValue)% total varians dalam data asli yang dapat dijelaskan oleh seluruh komponen utama. | |
| 141 | + | Total = sum(eValue); % total varians dalam data asli yang dapat dijelaskan oleh seluruh komponen utama. | |
| 139 | 142 | ||
| 140 | 143 | % Proporsi kumulatif = mengambil data seberapa persen secara kumulatif yang | |
| 141 | 144 | % dianggap perlu | |
| 142 | 145 | PK = 0; | |
| 143 | - | komponen = 0;%jumlah komponen utama yang dipertahankan | |
| 146 | + | komponen = 0; % jumlah komponen utama yang dipertahankan | |
| 144 | 147 | ||
| 148 | + | proporsiValue = zeros(p, 1); | |
| 145 | 149 | for j = 1:p | |
| 146 | - | proporsiValue(j,:) = eValue(j,1) / Total;%mengukur seberapa besar kontribusi eigenvalue ke-j terhadap varians total data. | |
| 150 | + | proporsiValue(j) = eValue(j,1) / Total; % mengukur seberapa besar kontribusi eigenvalue ke-j terhadap varians total data. | |
| 147 | 151 | if(PK < 0.9) % Nilai PK tergantung peneliti --> 0.8 | |
| 148 | - | PK = PK + proporsiValue(j, :); | |
| 152 | + | PK = PK + proporsiValue(j); | |
| 149 | 153 | komponen = komponen + 1; | |
| 150 | 154 | end | |
| 151 | 155 | end | |
| 152 | 156 | ||
| 153 | 157 | % Mengambil eigen vector dari komponen yang terpilih | |
| 154 | - | VA_Komponen = VA_urut(:,1:komponen) | |
| 158 | + | VA_Komponen = VA_urut(:,1:komponen); | |
| 155 | 159 | % W = variable komponen utama / principal component variable | |
| 156 | - | W = Z * VA_Komponen % W ADALAH FINAL DATA YANG DISEBUTKAN DI PDF | |
| 160 | + | W = Zr * VA_Komponen; % W ADALAH FINAL DATA YANG DISEBUTKAN DI PDF | |
| 157 | 161 | ||
| 158 | - | plot3(W(:,1), W(:,2), W(:,3), '*') % PK < 0.9 | |
| 159 | - | %plot(W(:,1), W(:,2), '*') % PK < 0.8 | |
| 162 | + | plot3(W(:,1), W(:,2), W(:,3), '*'); % PK < 0.9 | |
| 163 | + | %plot(W(:,1), W(:,2), '*'); % PK < 0.8 | |
| 160 | 164 | grid on | |
| @@ -0,0 +1 @@ | |||
| 1 | + | . | |
| @@ -0,0 +1,160 @@ | |||
| 1 | + | clear; clc; | |
| 2 | + | ||
| 3 | + | % Get Data | |
| 4 | + | gbr1 = imread('images/Banana/1.jpg'); | |
| 5 | + | gbr2 = imread('images/Banana/2.jpg'); | |
| 6 | + | gbr3 = imread('images/Banana/3.jpg'); | |
| 7 | + | gbr4 = imread('images/Banana/4.jpg'); | |
| 8 | + | gbr5 = imread('images/Banana/5.jpg'); | |
| 9 | + | gbr6 = imread('images/Banana/6.jpg'); | |
| 10 | + | gbr7 = imread('images/Banana/7.jpg'); | |
| 11 | + | gbr8 = imread('images/Banana/8.jpg'); | |
| 12 | + | gbr9 = imread('images/Banana/9.jpg'); | |
| 13 | + | gbr10 = imread('images/Banana/10.jpg'); | |
| 14 | + | gbr11 = imread('images/Kiwi/1.jpg'); | |
| 15 | + | gbr12 = imread('images/Kiwi/2.jpg'); | |
| 16 | + | gbr13 = imread('images/Kiwi/3.jpg'); | |
| 17 | + | gbr14 = imread('images/Kiwi/4.jpg'); | |
| 18 | + | gbr15 = imread('images/Kiwi/5.jpg'); | |
| 19 | + | gbr16 = imread('images/Kiwi/6.jpg'); | |
| 20 | + | gbr17 = imread('images/Kiwi/7.jpg'); | |
| 21 | + | gbr18 = imread('images/Kiwi/8.jpg'); | |
| 22 | + | gbr19 = imread('images/Kiwi/9.jpg'); | |
| 23 | + | gbr20 = imread('images/Kiwi/10.jpg'); | |
| 24 | + | gbr21 = imread('images/Orange/1.jpg'); | |
| 25 | + | gbr22 = imread('images/Orange/2.jpg'); | |
| 26 | + | gbr23 = imread('images/Orange/3.jpg'); | |
| 27 | + | gbr24 = imread('images/Orange/4.jpg'); | |
| 28 | + | gbr25 = imread('images/Orange/5.jpg'); | |
| 29 | + | gbr26 = imread('images/Orange/6.jpg'); | |
| 30 | + | gbr27 = imread('images/Orange/7.jpg'); | |
| 31 | + | gbr28 = imread('images/Orange/8.jpg'); | |
| 32 | + | gbr29 = imread('images/Orange/9.jpg'); | |
| 33 | + | gbr30 = imread('images/Orange/10.jpg'); | |
| 34 | + | gbr1=double(rgb2gray(gbr1)); | |
| 35 | + | gbr2=double(rgb2gray(gbr2)); | |
| 36 | + | gbr3=double(rgb2gray(gbr3)); | |
| 37 | + | gbr4=double(rgb2gray(gbr4)); | |
| 38 | + | gbr5=double(rgb2gray(gbr5)); | |
| 39 | + | gbr6=double(rgb2gray(gbr6)); | |
| 40 | + | gbr7=double(rgb2gray(gbr7)); | |
| 41 | + | gbr8=double(rgb2gray(gbr8)); | |
| 42 | + | gbr9=double(rgb2gray(gbr9)); | |
| 43 | + | gbr10=double(rgb2gray(gbr10)); | |
| 44 | + | gbr11=double(rgb2gray(gbr11)); | |
| 45 | + | gbr12=double(rgb2gray(gbr12)); | |
| 46 | + | gbr13=double(rgb2gray(gbr13)); | |
| 47 | + | gbr14=double(rgb2gray(gbr14)); | |
| 48 | + | gbr15=double(rgb2gray(gbr15)); | |
| 49 | + | gbr16=double(rgb2gray(gbr16)); | |
| 50 | + | gbr17=double(rgb2gray(gbr17)); | |
| 51 | + | gbr18=double(rgb2gray(gbr18)); | |
| 52 | + | gbr19=double(rgb2gray(gbr19)); | |
| 53 | + | gbr20=double(rgb2gray(gbr20)); | |
| 54 | + | gbr21=double(rgb2gray(gbr21)); | |
| 55 | + | gbr22=double(rgb2gray(gbr22)); | |
| 56 | + | gbr23=double(rgb2gray(gbr23)); | |
| 57 | + | gbr24=double(rgb2gray(gbr24)); | |
| 58 | + | gbr25=double(rgb2gray(gbr25)); | |
| 59 | + | gbr26=double(rgb2gray(gbr26)); | |
| 60 | + | gbr27=double(rgb2gray(gbr27)); | |
| 61 | + | gbr28=double(rgb2gray(gbr28)); | |
| 62 | + | gbr29=double(rgb2gray(gbr29)); | |
| 63 | + | gbr30=double(rgb2gray(gbr30)); | |
| 64 | + | ||
| 65 | + | gbr1=reshape(gbr1',1,[]); | |
| 66 | + | gbr2=reshape(gbr2',1,[]); | |
| 67 | + | gbr3=reshape(gbr3',1,[]); | |
| 68 | + | gbr4=reshape(gbr4',1,[]); | |
| 69 | + | gbr5=reshape(gbr5',1,[]); | |
| 70 | + | gbr6=reshape(gbr6',1,[]); | |
| 71 | + | gbr7=reshape(gbr7',1,[]); | |
| 72 | + | gbr8=reshape(gbr8',1,[]); | |
| 73 | + | gbr9=reshape(gbr9',1,[]); | |
| 74 | + | gbr10=reshape(gbr10',1,[]); | |
| 75 | + | gbr11=reshape(gbr11',1,[]); | |
| 76 | + | gbr12=reshape(gbr12',1,[]); | |
| 77 | + | gbr13=reshape(gbr13',1,[]); | |
| 78 | + | gbr14=reshape(gbr14',1,[]); | |
| 79 | + | gbr15=reshape(gbr15',1,[]); | |
| 80 | + | gbr16=reshape(gbr16',1,[]); | |
| 81 | + | gbr17=reshape(gbr17',1,[]); | |
| 82 | + | gbr18=reshape(gbr18',1,[]); | |
| 83 | + | gbr19=reshape(gbr19',1,[]); | |
| 84 | + | gbr20=reshape(gbr20',1,[]); | |
| 85 | + | gbr21=reshape(gbr21',1,[]); | |
| 86 | + | gbr22=reshape(gbr22',1,[]); | |
| 87 | + | gbr23=reshape(gbr23',1,[]); | |
| 88 | + | gbr24=reshape(gbr24',1,[]); | |
| 89 | + | gbr25=reshape(gbr25',1,[]); | |
| 90 | + | gbr26=reshape(gbr26',1,[]); | |
| 91 | + | gbr27=reshape(gbr27',1,[]); | |
| 92 | + | gbr28=reshape(gbr28',1,[]); | |
| 93 | + | gbr29=reshape(gbr29',1,[]); | |
| 94 | + | gbr30=reshape(gbr30',1,[]); | |
| 95 | + | ||
| 96 | + | X = [gbr1, gbr2, gbr3, gbr4, gbr5, gbr6, gbr7, gbr8, gbr9, gbr10, gbr11, gbr12, gbr13, gbr14, gbr15, gbr16, gbr17, gbr18, gbr19, gbr20, gbr21, gbr22, gbr23, gbr24, gbr25, gbr26, gbr27, gbr28, gbr29, gbr30]; | |
| 97 | + | [n,p] = size(X); | |
| 98 | + | Xrata = mean(X) | |
| 99 | + | ||
| 100 | + | %1. mengurangi setiap variable dengan rata2nya (normalisasi data) | |
| 101 | + | for i = 1:n | |
| 102 | + | Z(i, :) = X(i, :)-Xrata; | |
| 103 | + | end | |
| 104 | + | ||
| 105 | + | %Menghitung covariance matrix | |
| 106 | + | S = cov(X); % varian | |
| 107 | + | ||
| 108 | + | %Standardization Data | |
| 109 | + | for i = 1:n | |
| 110 | + | for j = 1:p | |
| 111 | + | Zr(i,j) = Z(i,j) / sqrt(S(j,j)); | |
| 112 | + | end | |
| 113 | + | end | |
| 114 | + | %ZR ADALAH HASIL DARI NORMALISASI DATA X (DATA AWAL) | |
| 115 | + | ||
| 116 | + | %2. Menghitung cov dari data yang sudah direduksi | |
| 117 | + | Zcov = cov(Zr); % matrix covarian dari variable hasil normalisasi Zr | |
| 118 | + | ||
| 119 | + | %3. Menghitung eigen value dan eigen vector dari Zcov | |
| 120 | + | [VA, DA] = eig(Zcov); % VA = Eigen Vector ; DA = Eigen Value | |
| 121 | + | ||
| 122 | + | %4. menghitung komponen utama | |
| 123 | + | % Hasil library matlab eig() --> hasil = diagonal utamanya urut dari kecil ke besar | |
| 124 | + | % yang diperlukan dari besar ke kecil | |
| 125 | + | %mengurutkan hasil eigen (DESC) | |
| 126 | + | m = 1 | |
| 127 | + | for i = p:-1:1 | |
| 128 | + | DA_urut(m,m) = DA(i,i); | |
| 129 | + | VA_urut(m,:) = VA(i,:); | |
| 130 | + | m = m+1; | |
| 131 | + | end | |
| 132 | + | ||
| 133 | + | %menghitung komponen utama | |
| 134 | + | for i = 1:p | |
| 135 | + | eValue(i,1) = DA_urut(i,i); | |
| 136 | + | end | |
| 137 | + | ||
| 138 | + | Total = sum(eValue)% total varians dalam data asli yang dapat dijelaskan oleh seluruh komponen utama. | |
| 139 | + | ||
| 140 | + | % Proporsi kumulatif = mengambil data seberapa persen secara kumulatif yang | |
| 141 | + | % dianggap perlu | |
| 142 | + | PK = 0; | |
| 143 | + | komponen = 0;%jumlah komponen utama yang dipertahankan | |
| 144 | + | ||
| 145 | + | for j = 1:p | |
| 146 | + | proporsiValue(j,:) = eValue(j,1) / Total;%mengukur seberapa besar kontribusi eigenvalue ke-j terhadap varians total data. | |
| 147 | + | if(PK < 0.9) % Nilai PK tergantung peneliti --> 0.8 | |
| 148 | + | PK = PK + proporsiValue(j, :); | |
| 149 | + | komponen = komponen + 1; | |
| 150 | + | end | |
| 151 | + | end | |
| 152 | + | ||
| 153 | + | % Mengambil eigen vector dari komponen yang terpilih | |
| 154 | + | VA_Komponen = VA_urut(:,1:komponen) | |
| 155 | + | % W = variable komponen utama / principal component variable | |
| 156 | + | W = Z * VA_Komponen % W ADALAH FINAL DATA YANG DISEBUTKAN DI PDF | |
| 157 | + | ||
| 158 | + | plot3(W(:,1), W(:,2), W(:,3), '*') % PK < 0.9 | |
| 159 | + | %plot(W(:,1), W(:,2), '*') % PK < 0.8 | |
| 160 | + | grid on | |