ClusterDB.hpp

Go to the documentation of this file.

/*==========================================================================
 * Copyright (c) 2002 University of Massachusetts.  All Rights Reserved.
 *
 * Use of the Lemur Toolkit for Language Modeling and Information Retrieval
 * is subject to the terms of the software license set forth in the LICENSE
 * file included with this software, and also available at
 * http://www.lemurproject.org/license.html
 *
 *==========================================================================
 */

// David Fisher, Peter Amstutz
// init: 11/18/2002
#ifndef _CLUSTERDB_HPP
#define _CLUSTERDB_HPP
#include <stdexcept>
#include "common_headers.hpp"
#include "Index.hpp"
#include "IndexedReal.hpp"
#include "ClusterParam.hpp"
#include "ClusterFactory.hpp"
#include "SimFactory.hpp"
#include "ThresholdFcn.hpp"
namespace lemur
{
  namespace api
  {
    
    class ClusterDBError : public runtime_error
    {
    public:
      ClusterDBError(const string& s) : runtime_error(s) { };
    };


    class ClusterDB
    {
    public:
      ClusterDB(const Index *ind, 
                double threshold = 0.25,
                enum ClusterParam::simTypes simType = ClusterParam::COS,
                enum ClusterParam::clusterTypes clusterType=ClusterParam::CENTROID,
                enum ClusterParam::docModes docMode = ClusterParam::DMAX);

      virtual ~ClusterDB();

      virtual int countClusters() const = 0;

      virtual int maxID() const = 0;

      virtual vector<lemur::cluster::Cluster*> getDocCluster(lemur::api::DOCID_T docId) const = 0;

      virtual lemur::cluster::Cluster *getCluster(int clusterId) const = 0;

      virtual vector<int> getDocClusterId(lemur::api::DOCID_T docId) const = 0;

      virtual int addToCluster(lemur::api::DOCID_T docId, int clusterId, double score) = 0;

      virtual int addToCluster(lemur::api::DOCID_T docId, lemur::cluster::Cluster *cluster, double score) = 0;

      virtual int removeFromCluster(lemur::api::DOCID_T docId, int clusterID) = 0;
      virtual int deleteCluster(int clusterID) = 0;

      virtual int deleteCluster (lemur::cluster::Cluster *target) = 0;

      int addCluster(lemur::cluster::Cluster *oldCluster);

      virtual vector<int> splitCluster(int cid, int num = 2) = 0;

      virtual int mergeClusters(int cid1, int cid2) = 0;

      virtual void printClusters() const;

      virtual int cluster(lemur::api::DOCID_T docId);

      virtual int cluster(lemur::api::DOCID_T docId, double &finalScore);

      virtual string getKeyWords(int cid, int numTerms = 10) const;

    protected:
      const Index *index;
      int numDocs;
      int numTerms;
      double threshold; // fix this.
      mutable vector<lemur::cluster::Cluster *> clusters;
      const lemur::cluster::SimilarityMethod *sim;
      lemur::cluster::ClusterFactory *factory;
      virtual lemur::cluster::Cluster * newCluster() = 0;
      lemur::cluster::Cluster* allocateCluster(int clusterID) const;
      lemur::cluster::ThresholdFcn *thresh;
    };
  }
}

#endif

---