featurefilter.java

一个很好的LIBSVM的JAVA源码。对于要研究和改进SVM算法的学者。可以参考。来自数据挖掘工具YALE工具包。

/*
 *  YALE - Yet Another Learning Environment
 *  Copyright (C) 2001-2004
 *      Simon Fischer, Ralf Klinkenberg, Ingo Mierswa, 
 *          Katharina Morik, Oliver Ritthoff
 *      Artificial Intelligence Unit
 *      Computer Science Department
 *      University of Dortmund
 *      44221 Dortmund,  Germany
 *  email: [e-mail]
 *  web:   http://yale.cs.uni-dortmund.de/
 *
 *  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., 59 Temple Place, Suite 330, Boston, MA 02111-1307
 *  USA.
 */
package edu.udo.cs.yale.operator.preprocessing;

import edu.udo.cs.yale.operator.Operator;
import edu.udo.cs.yale.operator.IOObject;
import edu.udo.cs.yale.operator.OperatorException;
import edu.udo.cs.yale.example.ExampleSet;
import edu.udo.cs.yale.example.Attribute;
import edu.udo.cs.yale.tools.LogService;

/**
 *  This is an abstract superclass for feature filters. These operators
 *  take an <code>ExampleSet</code> as input and switch off all features that
 *  fulfill a condition. Subclasses must implement the condition in the method
 *  <code>switchOffFeature()</code>, which returns a boolean.
 *
 *  @version $Id: FeatureFilter.java,v 1.2 2004/08/27 11:57:44 ingomierswa Exp $
 */
public abstract class FeatureFilter extends Operator
{
    private static final Class[] INPUT_CLASSES = { ExampleSet.class };
    private static final Class[] OUTPUT_CLASSES = { ExampleSet.class };

    /**
     * Must be implemented by subclasses. When operators extending this class
     * are applied, they loop through the set of features and switch off all features
     * for which this method returns TRUE, while keeping the status of the other
     * features.
     *
     * @param theFeature Feature to check.
     * @return TRUE if this feature should <b>not</b> be active in the output example
     *          set of this operator. FALSE otherwise.
     */
    public abstract boolean switchOffFeature(Attribute theFeature) throws OperatorException;

    /**
     * Applies filtering of features by looping through all features and checking
     * <code>switchOffFeature()</code>. If TRUE is returned, the feature is
     * switched off, ie it won't be used by the following operators in the chain.
     * If FALSE is returned by <code>switchOffFeature()</code>, the feature will
     * keep its previous status.
     *
     * @return An array of IOObjects, with the output example set as the only member.
     */
    public IOObject[] apply() throws OperatorException {
	
	ExampleSet eSet = (ExampleSet)((ExampleSet)getInput(ExampleSet.class)).clone();

        LogService.logMessage("FeatureFilter: found " + eSet.getNumberOfAttributes() + " features.", LogService.STATUS);

        // loop through the features:
	int i = 0;
	while (i < eSet.getNumberOfAttributes()) {
	    Attribute attribute = eSet.getAttribute(i);
	    if (switchOffFeature(attribute)) {
                eSet.removeAttribute(attribute);
            } else {
		i++;
	    }
        }

        LogService.logMessage("FeatureFilter: " + eSet.getNumberOfAttributes() + 
			      " features left after filtering.", LogService.STATUS);

        return new IOObject[] { eSet }; // this is a clone!
    }

    public Class[] getInputClasses() {  return INPUT_CLASSES;  }
    public Class[] getOutputClasses() {  return OUTPUT_CLASSES;   }
}