Package org.apache.lucene.classification.utils

Class ConfusionMatrixGenerator.ConfusionMatrix

  • Enclosing class:
    ConfusionMatrixGenerator
    public static class ConfusionMatrixGenerator.ConfusionMatrix
extends java.lang.Object
    a confusion matrix, backed by a Map representing the linearized matrix

    • Constructor Summary

      Constructors 
      Modifier Constructor Description
      private ConfusionMatrix​(java.util.Map<java.lang.String,​java.util.Map<java.lang.String,​java.lang.Long>> linearizedMatrix, double avgClassificationTime, int numberOfEvaluatedDocs)  

    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      double getAccuracy()
      Calculate accuracy on this confusion matrix using the formula: accuracy = correctly-classified / (correctly-classified + wrongly-classified)
      double getAvgClassificationTime()
      get the average classification time in milliseconds
      double getF1Measure()
      get the F-1 measure on this confusion matrix
      double getF1Measure​(java.lang.String klass)
      get the F-1 measure of the given class
      java.util.Map<java.lang.String,​java.util.Map<java.lang.String,​java.lang.Long>> getLinearizedMatrix()
      get the linearized confusion matrix as a Map
      int getNumberOfEvaluatedDocs()
      get the no.
      double getPrecision()
      get the macro averaged precision (see getPrecision(String)) over all the classes.
      double getPrecision​(java.lang.String klass)
      calculate precision on the given class
      double getRecall()
      get the macro averaged recall (see getRecall(String)) over all the classes
      double getRecall​(java.lang.String klass)
      calculate recall on the given class
      java.lang.String toString()  

      • Methods inherited from class java.lang.Object

        clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

    • Field Detail

      • linearizedMatrix

        private final java.util.Map<java.lang.String,​java.util.Map<java.lang.String,​java.lang.Long>> linearizedMatrix

      • avgClassificationTime

        private final double avgClassificationTime

      • numberOfEvaluatedDocs

        private final int numberOfEvaluatedDocs

      • accuracy

        private double accuracy

    • Constructor Detail

      • ConfusionMatrix

        private ConfusionMatrix​(java.util.Map<java.lang.String,​java.util.Map<java.lang.String,​java.lang.Long>> linearizedMatrix,
                        double avgClassificationTime,
                        int numberOfEvaluatedDocs)

    • Method Detail

      • getLinearizedMatrix

        public java.util.Map<java.lang.String,​java.util.Map<java.lang.String,​java.lang.Long>> getLinearizedMatrix()
        get the linearized confusion matrix as a Map
        Returns:
        a Map whose keys are the correct classification answers and whose values are the actual answers' counts

      • getPrecision

        public double getPrecision​(java.lang.String klass)
        calculate precision on the given class
        Parameters:
        klass - the class to calculate the precision for
        Returns:
        the precision for the given class

      • getRecall

        public double getRecall​(java.lang.String klass)
        calculate recall on the given class
        Parameters:
        klass - the class to calculate the recall for
        Returns:
        the recall for the given class

      • getF1Measure

        public double getF1Measure​(java.lang.String klass)
        get the F-1 measure of the given class
        Parameters:
        klass - the class to calculate the F-1 measure for
        Returns:
        the F-1 measure for the given class

      • getF1Measure

        public double getF1Measure()
        get the F-1 measure on this confusion matrix
        Returns:
        the F-1 measure

      • getAccuracy

        public double getAccuracy()
        Calculate accuracy on this confusion matrix using the formula: accuracy = correctly-classified / (correctly-classified + wrongly-classified)
        Returns:
        the accuracy

      • getPrecision

        public double getPrecision()
        get the macro averaged precision (see getPrecision(String)) over all the classes.
        Returns:
        the macro averaged precision as computed from the confusion matrix

      • getRecall

        public double getRecall()
        get the macro averaged recall (see getRecall(String)) over all the classes
        Returns:
        the recall as computed from the confusion matrix

      • toString

        public java.lang.String toString()
        Overrides:
        toString in class java.lang.Object

      • getAvgClassificationTime

        public double getAvgClassificationTime()
        get the average classification time in milliseconds
        Returns:
        the avg classification time

      • getNumberOfEvaluatedDocs

        public int getNumberOfEvaluatedDocs()
        get the no. of documents evaluated while generating this confusion matrix
        Returns:
        the no. of documents evaluated