timeseriesdeltafilter.java

一个数据挖掘系统的源码

/**
 *
 *   AgentAcademy - an open source Data Mining framework for
 *   training intelligent agents
 *
 *   Copyright (C)   2001-2003 AA Consortium.
 *
 *   This library is open source 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 2.0 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 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
 *
 */

package org.agentacademy.modules.dataminer.filters;

/**
 * <p>Title: The Data Miner prototype</p>
 * <p>Description: A prototype for the DataMiner (DM), the Agent Academy (AA) module responsible for performing data mining on the contents of the Agent Use Repository (AUR). The extracted knowledge is to be sent back to the AUR in the form of a PMML document.</p>
 * <p>Copyright: Copyright (c) 2002</p>
 * <p>Company: CERTH</p>
 * @author asymeon
 * @version 0.3
 */

import org.agentacademy.modules.dataminer.core.Instance;
import org.agentacademy.modules.dataminer.core.Instances;
import org.agentacademy.modules.dataminer.core.SparseInstance;
import org.agentacademy.modules.dataminer.core.UnsupportedAttributeTypeException;
import org.agentacademy.modules.dataminer.core.Utils;
import org.apache.log4j.Logger;
/**
 * An instance filter that assumes instances form time-series data and
 * replaces attribute values in the current instance with the difference
 * between the current value and the equivalent attribute attribute value
 * of some previous (or future) instance. For instances where the time-shifted
 * value is unknown either the instance may be dropped, or missing values
 * used.<p>
 *
 * Valid filter-specific options are:<p>
 *
 * -R index1,index2-index4,...<br>
 * Specify list of columns to calculate new values for.
 * First and last are valid indexes.
 * (default none)<p>
 *
 * -V <br>
 * Invert matching sense (i.e. calculate for all non-specified columns)<p>
 *
 * -I num <br>
 * The number of instances forward to take value differences between.
 * A negative number indicates taking values from a past instance.
 * (default -1) <p>
 *
 * -M <br>
 * For instances at the beginning or end of the dataset where the delta
 * values are not known, use missing values (default is to remove those
 * instances). <p>
 *
 */
public class TimeSeriesDeltaFilter extends TimeSeriesTranslateFilter {
  public static Logger                log = Logger.getLogger(TimeSeriesDeltaFilter.class);
  /**
   * Sets the format of the input instances.
   *
   * @param instanceInfo an Instances object containing the input instance
   * structure (any instances contained in the object are ignored - only the
   * structure is required).
   * @return true if the outputFormat may be collected immediately
   * @exception UnsupportedAttributeTypeException if selected
   * attributes are not numeric.
   */
  public boolean setInputFormat(Instances instanceInfo) throws Exception {

    super.setInputFormat(instanceInfo);
    // Create the output buffer
    Instances outputFormat = new Instances(instanceInfo, 0);
    for(int i = 0; i < instanceInfo.numAttributes(); i++) {
      if (m_SelectedCols.isInRange(i)) {
	if (outputFormat.attribute(i).isNumeric()) {
	  outputFormat.renameAttribute(i, outputFormat.attribute(i).name()
				       + " d"
				       + (m_InstanceRange < 0 ? '-' : '+')
				       + Math.abs(m_InstanceRange));
	} else {
	  throw new UnsupportedAttributeTypeException("Time delta attributes must be numeric!");
	}
      }
    }
    setOutputFormat(outputFormat);
    return true;
  }


  /**
   * Creates a new instance the same as one instance (the "destination")
   * but with some attribute values copied from another instance
   * (the "source")
   *
   * @param source the source instance
   * @param dest the destination instance
   * @return the new merged instance
   * @exception Exception if a problem occurs during merging
   */
  protected Instance mergeInstances(Instance source, Instance dest) {

    Instances outputFormat = outputFormatPeek();
    double[] vals = new double[outputFormat.numAttributes()];
    for(int i = 0; i < vals.length; i++) {
      if (m_SelectedCols.isInRange(i)) {
	if ((source != null)
	    && !source.isMissing(i)
	    && !dest.isMissing(i)) {
	  vals[i] = dest.value(i) - source.value(i);
	}
      } else {
	vals[i] = dest.value(i);
      }
    }
    Instance inst = null;
    if (dest instanceof SparseInstance) {
      inst = new SparseInstance(dest.weight(), vals);
    } else {
      inst = new Instance(dest.weight(), vals);
    }
    inst.setDataset(dest.dataset());
    return inst;
  }

  /**
   * Main method for testing this class.
   *
   * @param argv should contain arguments to the filter: use -h for help
   */
  public static void main(String [] argv) {

    try {
      if (Utils.getFlag('b', argv)) {
 	Filter.batchFilterFile(new TimeSeriesDeltaFilter(), argv);
      } else {
	Filter.filterFile(new TimeSeriesDeltaFilter(), argv);
      }
    } catch (Exception ex) {
      log.error(ex.getMessage());
    }
  }
}