public class ImpreciseEntropyContingencyTables extends ContingencyTables
This class inherits from the weka.core.ContingencyTables class of weka; The added functionality is for implementing a new entropy measure based on imprecise probabilities and uncertainty measures.
For more info:
J. Abellán, S. Moral Building classification trees using the total uncertainty criterion International Journal of Intelligent Systems, 18 (12) (2003), pp. 1215–1225 http://dx.doi.org/10.1002/int.10143
log2
Constructor and Description |
---|
ImpreciseEntropyContingencyTables() |
Modifier and Type | Method and Description |
---|---|
static double |
impreciseEntropy(double[] array,
double svalue)
Computes the entropy based on imprecise probabilities of the
given array.
|
static double |
impreciseEntropyConditionedOnRows(double[][] matrix,
double svalue)
Computes conditional entropy (based on imprecise probabilities) of the
columns given the rows.
|
static double |
impreciseEntropyOverColumns(double[][] matrix,
double svalue)
Computes the columns' entropy (based on imprecise probabilities) for the
given contingency table.
|
static void |
main(java.lang.String[] args)
Main method for testing this class.
|
static double[] |
sDistribution(double[] array,
int n,
double svalue)
This method distributes the mass of "s" to calculate the maximum entropy.
|
chiSquared, chiVal, cochransCriterion, CramersV, entropy, entropyConditionedOnColumns, entropyConditionedOnRows, entropyConditionedOnRows, entropyOverColumns, entropyOverRows, gainRatio, getRevision, lnFunc, log2MultipleHypergeometric, reduceMatrix, symmetricalUncertainty, tauVal
public static double[] sDistribution(double[] array, int n, double svalue)
array
- the arrayn
- the array effective sizesvalue
- s parameter used for the IDM (Imprecise Dirichlet Model).
The IDM is used to obtain imprecise probabilities as
intervals.public static double impreciseEntropy(double[] array, double svalue)
array
- the arraysvalue
- s parameter used for the IDM (Imprecise Dirchlet Model).
The IDM is used to obtain imprecise probabilities as
intervals.public static double impreciseEntropyOverColumns(double[][] matrix, double svalue)
matrix
- the contingency tablesvalue
- s parameter used for the IDM (Imprecise Dirchlet Model).
The IDM is used to obtain imprecise probabilities as
intervals.public static double impreciseEntropyConditionedOnRows(double[][] matrix, double svalue)
matrix
- the contingency table.svalue
- s parameter used for the IDM (Imprecise Dirchlet Model).
The IDM is used to obtain imprecise probabilities as
intervals.public static void main(java.lang.String[] args)
args
- the commandline options