FeatureWeighting.java example

Explorer
ComplexRapidMiner-master
- operator
- src
/*
 *  RapidMiner
 *
 *  Copyright (C) 2001-2008 by Rapid-I and the contributors
 *
 *  Complete list of developers available at our web site:
 *
 *       http://rapid-i.com
 *
 *  This program is free software: you can redistribute it and/or modify
 *  it under the terms of the GNU Affero General Public License as published by
 *  the Free Software Foundation, either version 3 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 Affero General Public License for more details.
 *
 *  You should have received a copy of the GNU Affero General Public License
 *  along with this program.  If not, see http://www.gnu.org/licenses/.
 */
package com.rapidminer.operator.features.weighting;

import java.util.LinkedList;
import java.util.List;

import com.rapidminer.example.ExampleSet;
import com.rapidminer.operator.IOObject;
import com.rapidminer.operator.OperatorDescription;
import com.rapidminer.operator.OperatorException;
import com.rapidminer.operator.features.FeatureOperator;
import com.rapidminer.operator.features.KeepBest;
import com.rapidminer.operator.features.Population;
import com.rapidminer.operator.features.PopulationOperator;
import com.rapidminer.operator.features.RedundanceRemoval;
import com.rapidminer.parameter.ParameterType;
import com.rapidminer.parameter.ParameterTypeInt;
import com.rapidminer.parameter.ParameterTypeString;


/**
 * This operator performs the weighting under the naive assumption that the
 * features are independent from each other. Each attribute is weighted with a
 * linear search. This approach may deliver good results after short time if the
 * features indeed are not highly correlated. <br />
 * The ideas of forward selection and backward elimination can easily be used
 * for the weighting with help of a {@link SimpleWeighting}.
 * 
 * @author Ingo Mierswa
 * @version $Id: FeatureWeighting.java,v 1.14 2006/03/27 13:22:00 ingomierswa
 *          Exp $
 */
public abstract class FeatureWeighting extends FeatureOperator {


	/** The parameter name for "Keep the best n individuals in each generation." */
	public static final String PARAMETER_KEEP_BEST = "keep_best";

	/** The parameter name for "Stop after n generations without improval of the performance." */
	public static final String PARAMETER_GENERATIONS_WITHOUT_IMPROVAL = "generations_without_improval";

	/** The parameter name for "Use these weights for the creation of individuals in each generation." */
	public static final String PARAMETER_WEIGHTS = "weights";
	private List<PopulationOperator> preOps = new LinkedList<PopulationOperator>();

	private List<PopulationOperator> postOps = new LinkedList<PopulationOperator>();

	private int generationsWOImp = 0;

	public abstract PopulationOperator getWeightingOperator(String parameter);

	public FeatureWeighting(OperatorDescription description) {
		super(description);
	}

	public IOObject[] apply() throws OperatorException {
		generationsWOImp = getParameterAsInt(PARAMETER_GENERATIONS_WITHOUT_IMPROVAL);

		preOps = new LinkedList<PopulationOperator>();
		preOps.add(new KeepBest(getParameterAsInt(PARAMETER_KEEP_BEST)));
		preOps.add(getWeightingOperator(getParameterAsString(PARAMETER_WEIGHTS)));
		preOps.add(new RedundanceRemoval());

		postOps = new LinkedList<PopulationOperator>();

		return super.apply();
	}

	public boolean solutionGoodEnough(Population population) {
		boolean stop = population.empty() || (population.getGenerationsWithoutImproval() >= generationsWOImp);
		return stop;
	}

	public List<PopulationOperator> getPreEvaluationPopulationOperators(ExampleSet eSet) {
		return preOps;
	}

	public List<PopulationOperator> getPostEvaluationPopulationOperators(ExampleSet eSet) {
		return postOps;
	}

	public List<ParameterType> getParameterTypes() {
		List<ParameterType> types = super.getParameterTypes();
		ParameterType type = new ParameterTypeInt(PARAMETER_KEEP_BEST, "Keep the best n individuals in each generation.", 1, Integer.MAX_VALUE, 1);
		type.setExpert(false);
		types.add(type);
		type = new ParameterTypeInt(PARAMETER_GENERATIONS_WITHOUT_IMPROVAL, "Stop after n generations without improval of the performance.", 1, Integer.MAX_VALUE, 1);
		type.setExpert(false);
		types.add(type);
		types.add(new ParameterTypeString(PARAMETER_WEIGHTS, "Use these weights for the creation of individuals in each generation.", true));
		return types;
	}
}