Class ClusterCenters

  • All Implemented Interfaces:
    AdditionalInformationHandler, Destroyable, GlobalInfoSupporter, LoggingLevelHandler, LoggingSupporter, OptionHandler, SizeOfHandler, Serializable

    public class ClusterCenters
    extends AbstractClusterMembershipPostProcessor
    Computes the cluster centers for the provided dataset. An additional attribute is added to the dataset structure that contains the cluster index.
    For numeric attributes the mean is calculated (or median, if 'useMedian' is enabled). For nominal attributes, the most common value is used.
    Stored in container under: Clustered dataset

    -logging-level <OFF|SEVERE|WARNING|INFO|CONFIG|FINE|FINER|FINEST> (property: loggingLevel)
        The logging level for outputting errors and debugging output.
        default: WARNING
    -use-median <boolean> (property: useMedian)
        If enabled, the median instead of the mean is calculated for numeric attributes.
        default: false
    fracpete (fracpete at waikato dot ac dot nz)
    See Also:
    Serialized Form
    • Field Detail

      • m_UseMedian

        protected boolean m_UseMedian
        compute median instead of mean.
    • Constructor Detail

      • ClusterCenters

        public ClusterCenters()
    • Method Detail

      • setUseMedian

        public void setUseMedian​(boolean value)
        Sets whether to use the median instead of the mean.
        value - true if to use median
      • getUseMedian

        public boolean getUseMedian()
        Returns whether to use the median instead of the mean.
        true if to use median
      • useMedianTipText

        public String useMedianTipText()
        Returns the tip text for this property.
        tip text for this property suitable for displaying in the GUI or for listing the options.
      • createOutputFormat

        protected weka.core.Instances createOutputFormat​(weka.core.Instances data)
        Generates the output format (additional attribute for cluster index).
        data - the original input data
        the header of the output format
      • calculateCenters

        protected weka.core.Instances calculateCenters​(weka.core.Instances data,
                                                       weka.clusterers.Clusterer clusterer,
                                                       weka.core.Instances outputFormat)
        Calculates the centers.
        data - the input data used for training the clusterer
        clusterer - the built clusterer
        outputFormat - the format to use for the output
        the generated output