clustboost.cpp

聚类分析程序 k-means 编译环境 gcc/stl

/*
  
  Text Clustering  Copyright (C) 2004 Debora "Barbara" Donato, Antonio Gulli

  This library is free software; you can redistribute it and/or modify it 
  under the terms of the GNU Lesser General Public License as published by 
  the Free Software Foundation; either version 2.1 of the License, or 
  (at your option) any later version.

  This library is distributed in the hope that it will be useful, but 
  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY 
  or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public 
  License for more details.

  You should have received a copy of the GNU Lesser General Public License 
  along with this library; if not, write to the Free Software Foundation, Inc., 
  59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 

*/
#include "dictionary.h"
#include "docvectors.h"
#include "k-mean-clusters.h"
#include "shingle-clustering.h"
#include "Reader.h"
#include "Clustering.h"
#include <stdio.h>
#include <getopt.h>

void test1(){

  vectorSpace vs1;
  document d_temp0;
  document d_temp1;
  document d_temp2;
  DOCUMENT::iterator d_temp_iter;
  double distance;
  //vector<tuple *> nums;
  //vector<tuple *>::iterator nums_iter;  
  
  //nums.push_back (new tuple(1,4));
  //nums.push_back (new tuple(2,7));
  //nums.push_back (new tuple(10,10));
  
  d_temp0.add(new tuple(1,4));
  d_temp0.add(new tuple(2,7));
  d_temp0.add(new tuple(10,10));
  
  cout << endl << "Vector 'd_temp0' now becomes:" << endl;
  for (d_temp_iter=d_temp0.begin(); d_temp_iter!=d_temp0.end(); d_temp_iter++){
    cout << (*d_temp_iter) << endl;
  }
  cout << endl; 
  
  d_temp1.add(new tuple(2,4));
  d_temp1.add(new tuple(3,7));

  cout << endl << "Vector 'd_temp1' now becomes:" << endl;
  for (d_temp_iter=d_temp1.begin(); d_temp_iter!=d_temp1.end(); d_temp_iter++){
    cout << (*d_temp_iter) << endl;
  }
  cout << endl; 

  d_temp2.add(new tuple(1,5));
  d_temp2.add(new tuple(3,3));
  d_temp2.add(new tuple(7,2));
  
  cout << endl << "Vector 'd_temp2' now becomes:" << endl;
  for (d_temp_iter=d_temp2.begin(); d_temp_iter!=d_temp2.end(); d_temp_iter++){
    cout << (*d_temp_iter) << endl;
  }
  cout << endl; 
 
 vs1.pushVector(d_temp0);
 vs1.pushVector(d_temp1);
 vs1.pushVector(d_temp2);
 
 distance = d_temp0.distance(d_temp1);
 cout << endl << "Distance (0,1):" << endl;
 cout << distance << endl;
 cout << endl;

 distance = d_temp1.distance(d_temp2);
 cout << endl << "Distance (1,2):" << endl;
 cout << distance << endl;
 cout << endl;

 distance = d_temp0.distance(d_temp2);
 cout << endl << "Distance (0,2):" << endl;
 cout << distance << endl;
 cout << endl;
 
 vector<kmCluster> clusters;                 // a collection of clusters
 kmCluster * a_cluster;                      // a cluster ptr
 a_cluster = new kmCluster();                // create new cluster
 
 a_cluster->addDocument(&d_temp0);
 a_cluster->addDocument(&d_temp1);
 a_cluster->addDocument(&d_temp2);
 
 a_cluster->do_centroid();
 cout << "Centroid is " << a_cluster->getCentroid();
  
}

inline void usage(){
  cerr << "Usage: test -k[kmeans]|-s[shingle]" <<endl;
  exit(0);
}

int main(int argc, char** argv){

  
  int option_char;
  bool shingleStart, kmeansStart;

  shingleStart = kmeansStart = false;

  while((option_char = getopt(argc, argv, "sk")) != -1){
   
    switch(option_char){
    case 's' :shingleStart = true; break;
    case 'k' : kmeansStart = true; break;
    case '?' : usage(); break;
    default:  break;
    }
  }
 if ((shingleStart && kmeansStart) || (!shingleStart && !kmeansStart))
   { usage();}  
 
 Clustering * clusterAlgo;
 
 if (shingleStart){
   
   clusterAlgo  = new SHINGLES(); // invoke the cluster algo   
   cout << "Shingling"<< endl;
 } else {
   
   clusterAlgo  = new KMEANS();    // invoke the cluster algo
   cout << "Kmeaning"<< endl;
 }
   
  Reader * reader  = new simpleReader();      // a simple reader 

  clusterAlgo->initialize();

  reader->loadDataSet(clusterAlgo);           // load the dataset

  clusterAlgo->do_clustering();                // do the clustering



}