cfo.m

function [predict_labels, U]=cfo(dataset, number_of_clusters,algorithm,seed)
    U = [];
    switch algorithm
        case 'kmeans'
%             [predict_labels,~] = kmeans(dataset,number_of_clusters);
                [theta]=rand_data_init(dataset',number_of_clusters,seed);
                [~,predict_labels,~]=k_means(dataset',theta);
        case 'fuzzy'
            options = [2.0 NaN NaN 0];
            [theta,U] = fcm(dataset,number_of_clusters, options);
            predict_labels = plot_fuzzy(number_of_clusters,theta,U,dataset);
        case 'possibilistic'
%             for q=1.5:1.0:2.5
                
            q = 1.7;
            eta = find_eta(dataset,number_of_clusters,q);
            [U,predict_labels]=possibi(dataset',number_of_clusters,eta,q,23,3,0.01);
%             tlt = "Possibilistic Clustering (q = " + q + ")";
%                 plot_possibilistic(number_of_clusters,predict_labels',dataset,tlt)
            predict_labels = plot_fuzzy(number_of_clusters,predict_labels',U',dataset);
%             view_image(predict_labels); 
%             end    
        case 'probabilistic'
            [theta]=rand_data_init(dataset',number_of_clusters,0);
            v = ones(3,1); 
            D = diag(v);
            for i=1:number_of_clusters
               cov(:,:,i) = D; 
            end 
            [~,cp,mv,~,~,~]=GMDAS(dataset',theta,cov,0.000001,1000,0);
%             plot_probabilistic(dataset, mv')
            predict_labels = plot_fuzzy(number_of_clusters,mv',cp',dataset);
                      
        otherwise
            disp('Error: There is no CFO algorithm with this name')
    end    
end