classificationrateassessment.java

一个数据挖掘软件ALPHAMINERR的整个过程的JAVA版源代码

/*
 *    This program is free software; you can redistribute it and/or modify
 *    it under the terms of the GNU General Public License as published by
 *    the Free Software Foundation; either version 2 of the License, or
 *    (at your option) any later version.
 *
 *    This program 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 General Public License for more details.
 *
 *    You should have received a copy of the GNU General Public License
 *    along with this program; if not, write to the Free Software
 *    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

/**
 * Title: XELOPES Data Mining Library
 * Description: The XELOPES library is an open platform-independent and data-source-independent library for Embedded Data Mining.
 * Copyright: Copyright (c) 2002 Prudential Systems Software GmbH
 * Company: ZSoft (www.zsoft.ru), Prudsys (www.prudsys.com)
 * @author Valentine Stepanenko ([e-mail])
 * @version 1.0
 */

package com.prudsys.pdm.Models.Classification;

import com.prudsys.pdm.Automat.ExternalDataModelAssessment;
import com.prudsys.pdm.Core.CategoricalAttribute;
import com.prudsys.pdm.Core.Category;
import com.prudsys.pdm.Core.MiningAttribute;
import com.prudsys.pdm.Core.MiningDataSpecification;
import com.prudsys.pdm.Core.MiningException;
import com.prudsys.pdm.Input.MiningVector;
import com.prudsys.pdm.Models.Supervised.SupervisedMiningModel;
import com.prudsys.pdm.Models.Supervised.SupervisedMiningSettings;

/**
 * Evaluates percentage of correctly classified vectors of a data set
 * for a given classifier.
 *
 * Performs "hard classification".
 */
public class ClassificationRateAssessment extends ExternalDataModelAssessment
{
   /**
    * Empty constructor.
    */
   public ClassificationRateAssessment() {

   }

   /**
    * Calculates classification rate.
    *
    * @return classification rate in percentage promille
    * @exception MiningException if assessment could not be calculated
    */
   public double calculateAssessment() throws MiningException {

     if ( miningModel == null || !(miningModel instanceof SupervisedMiningModel) )
       throw new MiningException("No supervised mining model specified");
     if (assessmentData == null)
       throw new MiningException("No assessment data specified");

     // Initializations:
     MiningDataSpecification metaData = assessmentData.getMetaData();
     MiningAttribute targetAttribute  = ((SupervisedMiningSettings) miningModel.getMiningSettings()).getTarget();

     // Calculate classification rate:
     int i = 0;
     int wrong = 0;
     assessmentData.reset();
     while (assessmentData.next()) {
       // Make prediction:
       MiningVector vector = assessmentData.read();
       double predicted    = ((SupervisedMiningModel) miningModel).applyModelFunction(vector);
       Category predTarCat = ((CategoricalAttribute)targetAttribute).getCategory(predicted);

       // Output and stats:
       double realTarCat   = vector.getValue( targetAttribute.getName() );
       Category tarCat     = ((CategoricalAttribute) metaData.getMiningAttribute(
           targetAttribute.getName())).getCategory(realTarCat);

       if (predTarCat == null || ! predTarCat.equals(tarCat) )
         wrong = wrong + 1;
       i = i + 1;
     };

     double classRate = 100.0 - ((double) wrong / i)*100.0;

     return classRate;
   }
}