kmeans.java

这是一个我自己编写的JAVA程序

package kmeans;
import java.io.*;
/**
 * The kmeans function perfotms a K-means clustering of 18 modes into two
 * clusters.
 * @author liyun
 * @version v1.1
 */
public class Kmeans{
    /**
     *it means ture
     */
    private static int SUCCESS=1;
    private static int TRUE=1;
    private static int FALSE=0;
    private static int MAXVECTDIM=20;
    private static int MAXPATTERN=20;
    private static int MAXCLUSTER=10;
    /**
     * The input Matrix
     * @see
     */
    private  double    pattern[][] = new double[MAXPATTERN][MAXVECTDIM] ;
    /**The Cluster arrays
     */
    private  ACluster   Cluster[] = new ACluster[MAXCLUSTER];
    /**number of patterns
     */
    private  int       numPatterns;
    /** Number of dimensions in vector
     */
    private  int       sizeVector;
    /** Number of clusters
     */
    private  int       numClusters;
    
    /**
     * Creates a new instance of System
     */
    public Kmeans() {}
  /** initial the clusters
   */  
    public void initClusters(){
        System.out.println("Initial cluster centers:");
        for (int i=0; i<numClusters; i++) {
            Cluster[i]=new ACluster();
            for (int j=0; j<sizeVector; j++) {
                Cluster[i].center[j]=pattern[i][j];
            } /* endfor */
        } /* endfor */
        for (int i=0; i<numClusters; i++) {
            System.out.println("ClusterCenter["+i+"]="+Cluster[i].center[0]+"  "+Cluster[i].center[1]);
        }
    }
    /**
     * Run the Kmeans
     */
    public void runKmeans(){
        int converged=FALSE;
        int pass =1;
        while (converged==FALSE) {
            System.out.println("PASS= "+(pass++));
            distributeSamples();
            converged=calcNewClustCenters();
            showCluster();
        } /* endwhile */
    }
    
    private void distributeSamples(){
        int i,pat,clustid,memberIndex;
        //Clear membership list for all current clusters
        for (i=0; i<numClusters;i++){
            Cluster[i].numMembers=0;
        }
        for (pat=0; pat<numPatterns; pat++) {
            /*Find cluster center to which the pattern is closest*/
            clustid= findClosestCluster(pat);
            System.out.println("pattern "+pat+" belong to  " +"cluster["+clustid +"]");
            /*post this pattern to the cluster*/
            memberIndex=Cluster[clustid].numMembers;
            Cluster[clustid].member[memberIndex]=pat;
            Cluster[clustid].numMembers++;
        } /* endfor */
    }
    
    private int findClosestCluster(int pat){
        int i, clustID;
        double minDist, d;
        minDist =9.9e+99;
        clustID=-1;
        for (i=0; i<numClusters; i++) {
            d=eucNorm(pat,i);
            System.out.println("pattern "+pat+" to the cluster [" +i+"]'s distance is "+d+" !");
            if (d<minDist) {
                minDist=d;
                clustID=i;
            } /* endif */
        } /* endfor */
        if (clustID<0) {
            System.out.println("Aaargh");
            System.exit(0);
        } /* endif */
        return clustID;
    }
    /**  Calc Euclidean norm of vector difference between pattern vector, p, and cluster  center, c.
     */
    private double eucNorm(int p, int c){
        double dist=0,x=0;
        int i;
        System.out.println("Now calculate pattern "+p+" to"+" the cluster["+c+"['s distance:");
        for (i=0; i<sizeVector ;i++){
            x += (Cluster[c].center[i]-pattern[p][i])*(Cluster[c].center[i]-pattern[p][i]);
        } /* endfor */
        return dist=Math.sqrt(x);
    }
    
    private int loadPatterns(String fname) throws java.io.IOException {
        File InFilePtr=new File(fname);
        FileInputStream f1=new FileInputStream(InFilePtr);
        DataInputStream fd1=new DataInputStream(f1);
        int i,j;
        if(fd1!=null){
            numPatterns=fd1.readInt();
            sizeVector=fd1.readInt();
            numClusters=fd1.readInt();
            for(i=0;i<numPatterns;i++){
                for(j=0;j<numClusters;j++){
                    double x=fd1.readDouble();
                    pattern[i][j]=x;
                }
            }
        }
        fd1.close();
        f1.close();
        for (i=0;i<18;i++){
            System.out.println("Pattern:"+pattern[i][0] +"  " +  pattern[i][1]);
        }
        return SUCCESS;
    }
    /** calculate the new clustCenters
     * @see
     */
    public int calcNewClustCenters(){
        int k=0;
        int vectID;
        int ConvFlag=TRUE;
        double tmp[]=new double[MAXVECTDIM];
        for(int j=0;j<MAXVECTDIM;j++){
            tmp[j]=0.0;
        }
        for(int i=0;i<numClusters;i++){
            for(int j=0;j<Cluster[i].numMembers;j++){
                vectID=Cluster[i].member[j];
                for (k=0;k<sizeVector;k++){
                    tmp[k]+=pattern[vectID][k];
                }
            }
            for(k=0;k<sizeVector;k++){
                tmp[k]=tmp[k]/(Cluster[i].numMembers);
                
                if(tmp[k]!=Cluster[i].center[k])
                    ConvFlag=FALSE;
                Cluster[i].center[k]=tmp[k];
            }
        }
        return ConvFlag;
    }
    
    /**
     * show clusters
     */    
    public void showCluster(){
        int cl;
        for(cl=0;cl<numClusters;cl++){
            System.out.println("CLUSTER["+cl+"]: "+Cluster[cl].center[0]+"   "+Cluster[cl].center[1]);
            for(int i=0;i<Cluster[cl].numMembers;i++)
                System.out.println("the Patterns in the Cluster["+cl+"] "+" are: "+pattern[Cluster[cl].member[i]][0]+
                "  "+pattern[Cluster[cl].member[i]][1] );
        }
    }
    
    /**
     * save clusters
     * @param
     * @since
     */    
    public void saveCluster(String fname){}
    
    /**
     * main methoes
     * @param
     * @
     */    
    public static void main(String[ ] args)throws java.io.IOException{
        Kmeans kmean=new Kmeans();
        if(kmean.loadPatterns("D:/javasources/ann/Data.txt")!=SUCCESS){
            System.out.println("UNABLE TO READ PATTERN_FILE:");
            System.exit(0);
        }
        kmean.initClusters();
        kmean.runKmeans();
    }
}