请使用matlab语言编写一个可以计算聚类结果的DBI、CHI、DI、轮廓系数指标结果的function函数，输入参数仅有原始数据集矩阵和聚类结果索引向量

在MATLAB中，你可以创建一个名为`cluster_measures`的函数，用于计算DBI (Davies-Bouldin Index)、CHI (Calinski-Harabasz Index)、DI (Dunn Index)和轮廓系数的值。这些指标常用于评估聚类效果的质量。以下是该函数的一个简单实现：

function [dbi, chi, di, silhouette] = cluster_measures(data, labels)
% 数据集矩阵 data 和聚类结果索引向量 labels 作为输入

% 确保输入合法
if ~isnumeric(data) || size(data, 2) ~= size(labels, 1)
    error('Invalid input! Data should be a matrix and labels should match the number of rows in data.');
end

% 轮廓系数（Silhouette Coefficient）
silhouette_values = silhouette(data, labels);
silhouette = mean(silhouette_values);

% 计算DBI (Davies-Bouldin Index)
n_clusters = length(unique(labels));
dbi = zeros(1, n_clusters);
for i = 1:n_clusters
    centroids = mean(data(labels == i), 1);
    distances = pdist2(centroids, data(labels == i)); % 计算簇内距离
    inter_class_distances = max(distances, [], 2); % 最大对角线距离
    intra_class_distances = min(distances, [], 2); % 最小对角线距离
    dbi(i) = mean(inter_class_distances) / mean(intra_class_distances);
end
dbi = -sum(dbi) / n_clusters; % 取负数因为通常越低越好

% 计算CHI (Calinski-Harabasz Index)
chi = (n_clusters * trace(var(data, 0))) / sum((length(labels) - n_clusters) * var(data(labels == labels(1)), 0));

% 计算DI (Dunn Index)
di = Inf;
for i = 1:n_clusters
    [min_dist, min_idx] = min(pdist2(data(labels == i)));
    for j = i+1:n_clusters
        [max_dist, max_idx] = max(pdist2(data(labels == i), data(labels == j)));
        if max_dist > min_dist
            di = min(di, min_dist / max_dist);
        end
    end
end

% 返回结果
[~, ind] = sort([dbi, chi, di], 'descend');
disp(['Cluster Measures: DBI: ', num2str(dbi(ind)), ', CHI: ', num2str(chi), ', DI: ', num2str(di), ', Silhouette: ', num2str(silhouette)]);
end

在这个函数中，我们首先检查输入的有效性，然后依次计算每个指标。最后，我们将各项指标按照从高到低排序并显示结果。