Module es.upm.fi.cig.multictbnc
Package es.upm.fi.cig.multictbnc.conceptdriftdetection

Class ConceptDriftAdaptiveMethod

java.lang.Object
es.upm.fi.cig.multictbnc.conceptdriftdetection.ConceptDriftAdaptiveMethod
Direct Known Subclasses:
ConceptDriftGloballyAdaptiveMethod, ConceptDriftLocallyAdaptiveMethod
public abstract class ConceptDriftAdaptiveMethod extends Object
Abstract class representing a concept drift adaptive method. This class provides a framework for implementing adaptive methods to handle concept drifts in Multi-CTBNCs.

  • Constructor Details

    • ConceptDriftAdaptiveMethod

      public ConceptDriftAdaptiveMethod(ConceptDriftScore conceptDriftScore, List<String> namesVariables, double detectionThreshold, boolean showCharts, String title)
      Initializes the concept drift adaptive method with the specified parameters.
      Parameters:
      conceptDriftScore - scoring mechanism to detect concept drift
      namesVariables - list of variable names used in the model
      detectionThreshold - threshold for determining concept drifts
      showCharts - flag indicating whether to display drift detection charts
      title - title for the drift detection charts

  • Method Details

    • setUpLineChart

      protected abstract void setUpLineChart(String title, double detectionThreshold)
      Sets up line charts for visualizing concept drift detection results if the charts are enabled.
      Parameters:
      title - title for the drift detection charts
      detectionThreshold - threshold for determining concept drifts

    • adaptModel

      public abstract boolean adaptModel(MultiCTBNC<CPTNode,CIMNode> model, Dataset batchDataStream) throws ErroneousValueException
      Abstract method for adapting the model based on the provided data batch. Implementations should define how the model is updated in response to concept drifts.
      Parameters:
      model - Multi-CTBNC model to adapt
      batchDataStream - new data batch
      Returns:
      true if the model was adapted, false otherwise
      Throws:
      ErroneousValueException - if an error occurs during model adaptation

    • getResults

      public abstract String getResults()
      Returns a String describing the results of the last concept drift detection.
      Returns:
      String describing the concept drift detection results

    • getUpdatingTime

      public long getUpdatingTime()
      Returns the time taken for the last update of the model.
      Returns:
      time taken for updating the model

    • getLastChangedNodes

      public abstract List<Node> getLastChangedNodes()
      Retrieves the list of nodes that were last identified as having undergone concept drift.
      Returns:
      list of nodes that experienced concept drift in the most recent adaptation