Q型聚类分析matlab代码，附上对聚类结果可视化的代码

以下是基于距离矩阵的Q型聚类分析matlab代码：

% 首先需要准备好距离矩阵，这里假设距离矩阵为D
% D是一个n*n的矩阵，其中D(i,j)表示第i个样本和第j个样本之间的距离
n = size(D, 1);

% 初始化聚类结果，一开始每个样本都是一个簇
clusters = cell(n, 1);
for i = 1:n
    clusters{i} = i;
end

% 开始合并簇，直到只剩下一个簇
while length(clusters) > 1
    % 找到最小的距离和对应的两个簇
    minDist = inf;
    minI = 0;
    minJ = 0;
    for i = 1:length(clusters)
        for j = (i+1):length(clusters)
            dist = computeDistance(clusters{i}, clusters{j}, D);
            if dist < minDist
                minDist = dist;
                minI = i;
                minJ = j;
            end
        end
    end
    
    % 合并簇
    clusters{minI} = [clusters{minI}, clusters{minJ}];
    clusters(minJ) = [];
end

% 可视化聚类结果，这里假设样本的特征为X，共有m个特征
m = size(X, 2);
for i = 1:length(clusters{1})
    plot(X(clusters{1}(i), 1), X(clusters{1}(i), 2), 'ro');
    hold on;
end
for i = 1:length(clusters{2})
    plot(X(clusters{2}(i), 1), X(clusters{2}(i), 2), 'go');
    hold on;
end
% 依此类推...

function dist = computeDistance(cluster1, cluster2, D)
% 计算两个簇之间的距离，这里采用Q型距离
n1 = length(cluster1);
n2 = length(cluster2);
dist = 0;
for i = 1:n1
    for j = 1:n2
        dist = dist + D(cluster1(i), cluster2(j));
    end
end
dist = dist / (n1*n2);
end

以下是可视化聚类结果的代码：

% 假设clusters是聚类结果，其中每个元素是一个簇
% 假设样本的特征为X，共有m个特征
m = size(X, 2);
for i = 1:length(clusters)
    color = rand(1, 3); % 随机生成一个颜色
    for j = 1:length(clusters{i})
        plot(X(clusters{i}(j), 1), X(clusters{i}(j), 2), 'o', 'color', color);
        hold on;
    end
end