pichuli.m

模式识别中的聚类算法

clear all; clc;
X = [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20;
     0 1 0 1 2 1 2 3 6 7  8  6  7  8  9  7  8  9  8  9;
     0 0 1 1 1 2 2 2 6 6  6  7  7  7  7  8  8  8  9  9]';
K = 2;
one_set = [ ];
two_set = [ ];
flag = 0;

one_center1 = X(1,2:3);
two_center1 = X(2,2:3);

one_center = [0 0];
two_center = [0 0];

m = sum(one_center1 - one_center);
n = sum(two_center1 - two_center);

while (m >= 10^-4 || n >= 10^-4)
    if flag< 1000
    one_center = one_center1;
    two_center = two_center1;
    
    one_set = [];
    two_set = [];
    
    for i = 1:20
        Y = X(i,2:3);
        dist1 = dot(Y - one_center, Y - one_center);
        dist2 = dot(Y - two_center, Y - two_center);
        if (dist1 < dist2)
            one_set = [one_set;X(i,:)];
        else
            two_set = [two_set;X(i,:)];
        end
    end
    if (size(one_set,1) == 1)
        one = one_set;
    else
        p = sum(one_set);
        q = size(one_set,1);
        one = p(:,2:3)/q;
    end
    if (size(two_set,1) == 1)
        two = two_set;
    else
        p = sum(two_set);
        q = size(two_set,1);
        two = p(:,2:3)/q;
    end 
    
    one_center1 = one;
    two_center1 = two;
    
    m = sum(one_center1 - one_center);
    n = sum(two_center1 - two_center);
    flag = flag + 1;
    end
end
flag
one_set
two_set

return