DecisionClass
, Union
public abstract class ApproximatedEntity
extends java.lang.Object
Modifier and Type | Field | Description |
---|---|---|
static int |
NONNEGATIVE_CONFIRMATION |
Possible value of
confirmationCheckMode , implying that descriptions of granules (i.e., conjunctions of attribute-relation-value triplets)
having nonnegative confirmation are considered to be confirmatory. |
static int |
POSITIVE_CONFIRMATION |
Possible value of
confirmationCheckMode , implying that only descriptions of granules (i.e., conjunctions of attribute-relation-value triplets)
having strictly positive confirmation are considered to be confirmatory. |
Constructor | Description |
---|---|
ApproximatedEntity() |
Modifier and Type | Method | Description |
---|---|---|
java.lang.Object |
clone() |
Clones this approximated entity
|
boolean |
containsExample(int exampleNumber) |
Checks if example with given number, from memory container for which this entity has been created,
belongs to this entity.
|
double |
getAccuracyOfApproximation() |
Gets accuracy of approximation of this entity by the set of all active condition attributes.
|
double |
getAccuracyOfApproximation(byte[] attributesMask) |
Gets accuracy of approximation of this entity by the set of active condition attributes determined by given mask.
|
double |
getAccuracyOfApproximation(double variableConsistencyParameterValue) |
Gets accuracy of approximation of this entity by the set of all active condition attributes.
|
double |
getAccuracyOfApproximation(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets accuracy of approximation of this entity by the set of active condition attributes determined by given mask.
|
Field |
getBasicClass() |
Gets basic class for this entity.
|
java.util.HashSet<java.lang.Integer> |
getBNDRegionExamples() |
Gets hash set with (
Integer ) numbers of examples belonging to boundary (BND) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA). |
java.util.HashSet<java.lang.Integer> |
getBNDRegionExamples(byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to boundary (BND) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes. |
java.util.HashSet<java.lang.Integer> |
getBNDRegionExamples(double variableConsistencyParameterValue) |
Gets hash set with (
Integer ) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA). |
java.util.HashSet<java.lang.Integer> |
getBNDRegionExamples(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes. |
int[] |
getBoundary() |
Gets numbers of examples belonging to boundary of this entity.
|
int[] |
getBoundary(byte[] attributesMask) |
Gets numbers of examples belonging to boundary of this entity.
|
int[] |
getBoundary(double variableConsistencyParameterValue) |
Gets numbers of examples belonging to boundary of this entity.
|
int[] |
getBoundary(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets numbers of examples belonging to boundary of this entity.
|
Field |
getCharacteristicDecisionClass() |
Gets decision class (value of decision attribute) characteristic for this entity.
|
abstract int |
getDecisionAttributeNumber() |
Gets the number of decision attribute for which this approximated entity is defined
|
abstract java.lang.String |
getDescription() |
Gets text description of this entity, showing e.g. its basic class.
|
int[] |
getExamples() |
Gets numbers of examples belonging to this entity.
Because of efficiency, reference to the field of this class, not its copy, is returned. |
java.util.HashSet<java.lang.Integer> |
getExamplesSet() |
Gets hash set with numbers of all examples belonging to this entity
Because of efficiency, reference to the field of this class, not its copy, is returned. |
java.util.HashSet<java.lang.Integer> |
getInconsistentExamplesForExtendedLowerApproximation(double variableConsistencyParameterValue) |
Gets hash set of (
Integer ) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter value (variable consistency approach). |
java.util.HashSet<java.lang.Integer> |
getInconsistentExamplesForExtendedLowerApproximation(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets hash set of (
Integer ) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter value (variable consistency approach)
and for given mask for attributes. |
int[] |
getLowerApproximation() |
Gets numbers of examples belonging to lower approximation of this entity
Calls getLowerApproximation(byte[]) method and passes null as mask for attributes
to indicate that mask for attributes should not be used.Because of efficiency, reference to the field of this class, not its copy, is returned. |
int[] |
getLowerApproximation(byte[] attributesMask) |
Gets numbers of examples belonging to lower approximation of this entity.
Uses given mask for attributes or considers all attributes if mask is null . |
int[] |
getLowerApproximation(double variableConsistencyParameterValue) |
Gets numbers of examples belonging to lower approximation of this entity.
|
int[] |
getLowerApproximation(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets numbers of examples belonging to lower approximation of this entity.
|
MemoryContainer |
getMemoryContainer() |
Gets reference to memory container for which this entity is defined
|
java.util.HashSet<java.lang.Integer> |
getNEGRegionExamples() |
Gets hash set with (
Integer ) numbers of examples belonging to negative (NEG) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA). |
abstract java.util.HashSet<java.lang.Integer> |
getNEGRegionExamples(byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to negative (NEG) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes. |
java.util.HashSet<java.lang.Integer> |
getNEGRegionExamples(double variableConsistencyParameterValue) |
Gets hash set with (
Integer ) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA). |
abstract java.util.HashSet<java.lang.Integer> |
getNEGRegionExamples(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes. |
int[] |
getNeutralExamples() |
Gets numbers of examples neutral to this entity (i.e., neither belonging to this entity, nor not belonging to this entity).
Because of efficiency, reference to the field of this class, not its copy, is returned. |
java.util.HashSet<java.lang.Integer> |
getNeutralExamplesSet() |
Gets hash set with numbers of all examples neutral to this entity (i.e., neither belonging to this entity, nor not belonging to this entity).
|
java.util.HashSet<java.lang.Integer> |
getPOSRegionExamples() |
Gets hash set with (
Integer ) numbers of examples belonging to positive (POS) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA). |
java.util.HashSet<java.lang.Integer> |
getPOSRegionExamples(byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to positive (POS) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes. |
java.util.HashSet<java.lang.Integer> |
getPOSRegionExamples(double variableConsistencyParameterValue) |
Gets hash set with (
Integer ) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA). |
java.util.HashSet<java.lang.Integer> |
getPOSRegionExamples(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets hash set with (
Integer ) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes. |
double |
getQualityOfApproximation() |
Gets quality of approximation of this entity by the set of all active condition attributes.
|
double |
getQualityOfApproximation(byte[] attributesMask) |
Gets quality of approximation of this entity by the set of active condition attributes determined by given mask.
|
double |
getQualityOfApproximation(double variableConsistencyParameterValue) |
Gets quality of approximation of this entity by the set of all active condition attributes.
|
double |
getQualityOfApproximation(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets quality of approximation of this entity by the set of active condition attributes determined by given mask.
|
abstract double |
getTheMostRestrictiveVariableConsistencyParameterValue() |
Gets the most restrictive value of used variable consistency parameter
|
int[] |
getUpperApproximation() |
Gets numbers of examples belonging to upper approximation of this entity.
Because of efficiency, reference to the field of this class, not its copy, is returned. |
int[] |
getUpperApproximation(byte[] attributesMask) |
Gets numbers of examples belonging to upper approximation of this entity.
Uses given mask for attributes or considers all attributes if mask is null . |
int[] |
getUpperApproximation(double variableConsistencyParameterValue) |
Gets numbers of examples belonging to upper approximation of this entity.
|
int[] |
getUpperApproximation(double variableConsistencyParameterValue,
byte[] attributesMask) |
Gets numbers of examples belonging to upper approximation of this entity.
|
void |
setBoundary(int[] boundary,
double boundaryVCParameterValue,
byte[] boundaryAttributesMask) |
Sets boundary of this entity and all parameters describing conditions under which this boundary has been calculated
|
void |
setLowerApproximation(int[] lowerApproximation,
double lowerApproximationVCParameterValue,
byte[] lowerApproximationAttributesMask) |
Sets lower approximation of this entity and all parameters describing conditions under which this approximation has been calculated
|
void |
setUpperApproximation(int[] upperApproximation,
double upperApproximationVCParameterValue,
byte[] upperApproximationAttributesMask) |
Sets upper approximation of this entity and all parameters describing conditions under which this approximation has been calculated
|
int |
size() |
Gets quantity of examples belonging to this entity
|
abstract boolean |
variableConsistencyParameterValueIsTheMostRestrictiveOne(double variableConsistencyParameterValue) |
Tells if given variable consistency parameter value is the most restrictive one.
|
public static final int POSITIVE_CONFIRMATION
confirmationCheckMode
, implying that only descriptions of granules (i.e., conjunctions of attribute-relation-value triplets)
having strictly positive confirmation are considered to be confirmatory.public static final int NONNEGATIVE_CONFIRMATION
confirmationCheckMode
, implying that descriptions of granules (i.e., conjunctions of attribute-relation-value triplets)
having nonnegative confirmation are considered to be confirmatory.public int[] getExamples()
public java.util.HashSet<java.lang.Integer> getExamplesSet()
public int[] getNeutralExamples()
null
should be returned!public java.util.HashSet<java.lang.Integer> getNeutralExamplesSet()
null
should be returned!public int[] getLowerApproximation()
getLowerApproximation(byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.public int[] getLowerApproximation(byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.
This method is designed to facilitate reducts calculations.attributesMask
- Mask for attributes allowing lower approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating lower approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating lower approximation.InvalidValueException
- when length of the mask for attributes is different than quantity of attributes
in memory containerpublic int[] getLowerApproximation(double variableConsistencyParameterValue)
variableConsistencyParameterValue
- variable consistency parameter value for which lower approximation should be calculated (VC-IRSA / VC-DRSA)InvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic int[] getLowerApproximation(double variableConsistencyParameterValue, byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which lower approximation should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing lower approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating lower approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating lower approximation.InvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic int[] getUpperApproximation()
public int[] getUpperApproximation(byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.attributesMask
- Mask for attributes allowing upper approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating upper approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating upper approximation.InvalidValueException
- when length of the mask for attributes is different than quantity of attributes
in memory containerpublic int[] getUpperApproximation(double variableConsistencyParameterValue)
getUpperApproximation(double, byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.variableConsistencyParameterValue
- variable consistency parameter value for which upper approximation should be calculated (VC-IRSA / VC-DRSA)InvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic int[] getUpperApproximation(double variableConsistencyParameterValue, byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which upper approximation should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing upper approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating upper approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating upper approximation.InvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic int[] getBoundary()
getBoundary(byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.public int[] getBoundary(byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.attributesMask
- Mask for attributes allowing boundary calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating boundary. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating boundary.InvalidValueException
- when length of the mask for attributes is different than quantity of attributes
in memory containerpublic int[] getBoundary(double variableConsistencyParameterValue)
getBoundary(double, byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.variableConsistencyParameterValue
- variable consistency parameter value for which boundary should be calculated (VC-IRSA / VC-DRSA)InvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic int[] getBoundary(double variableConsistencyParameterValue, byte[] attributesMask)
null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which boundary should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing boundary calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating boundary. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating boundary.InvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic double getAccuracyOfApproximation()
getAccuracyOfApproximation(byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.public double getAccuracyOfApproximation(byte[] attributesMask)
attributesMask
- Mask for attributes allowing accuracy of approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating accuracy of approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating accuracy of approximation.InvalidValueException
- when length of the mask for attributes is different than quantity of attributes
in memory containerpublic double getAccuracyOfApproximation(double variableConsistencyParameterValue)
getAccuracyOfApproximation(double, byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.variableConsistencyParameterValue
- variable consistency parameter value for which accuracy of approximation should be calculated (VC-IRSA / VC-DRSA)public double getAccuracyOfApproximation(double variableConsistencyParameterValue, byte[] attributesMask)
variableConsistencyParameterValue
- variable consistency parameter value for which accuracy of approximation should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing accuracy of approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating accuracy of approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating accuracy of approximation.InvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic double getQualityOfApproximation()
getQualityOfApproximation(byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.public double getQualityOfApproximation(byte[] attributesMask)
attributesMask
- Mask for attributes allowing quality of approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating quality of approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating quality of approximation.InvalidValueException
- when length of the mask for attributes is different than quantity of attributes
in memory containerpublic double getQualityOfApproximation(double variableConsistencyParameterValue)
getQualityOfApproximation(double, byte[])
method and passes null
as mask for attributes
to indicate that mask for attributes should not be used.variableConsistencyParameterValue
- variable consistency parameter value for which quality of approximation should be calculated (VC-IRSA / VC-DRSA)public double getQualityOfApproximation(double variableConsistencyParameterValue, byte[] attributesMask)
variableConsistencyParameterValue
- variable consistency parameter value for which quality of approximation should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing quality of approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating quality of approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating quality of approximation.InvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic int size()
public MemoryContainer getMemoryContainer()
public Field getBasicClass()
public Field getCharacteristicDecisionClass()
public boolean containsExample(int exampleNumber)
exampleNumber
- number of example, concerning memory container for which this entity has been createdtrue
if this entity contains example (from used memory container) having given number,
false
otherwisepublic void setLowerApproximation(int[] lowerApproximation, double lowerApproximationVCParameterValue, byte[] lowerApproximationAttributesMask)
lowerApproximation
- array with numbers of objects from memory container which belong to lower approximation of this entitylowerApproximationVCParameterValue
- VC-IRSA / VC-DRSA parameter value for which given lower approximation has been calculatedlowerApproximationAttributesMask
- mask for attributes allowing lower approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute has not been taken into account
at all when calculating lower approximation. If attributesMask(i) is different than zero,
than corresponding attribute has been taken into account when calculating lower approximation.InvalidValueException
- when VC-IRSA / VC-DRSA parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic void setUpperApproximation(int[] upperApproximation, double upperApproximationVCParameterValue, byte[] upperApproximationAttributesMask)
upperApproximation
- array with numbers of objects from memory container which belong to upper approximation of this entityupperApproximationVCParameterValue
- VC-IRSA / VC-DRSA parameter value for which given upper approximation has been calculatedupperApproximationAttributesMask
- mask for attributes allowing upper approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute has not been taken into account
at all when calculating upper approximation. If attributesMask(i) is different than zero,
than corresponding attribute has been taken into account when calculating upper approximation.InvalidValueException
- when VC-IRSA / VC-DRSA parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic void setBoundary(int[] boundary, double boundaryVCParameterValue, byte[] boundaryAttributesMask)
boundary
- array with numbers of objects from memory container which belong to boundary of this entityboundaryVCParameterValue
- VC-IRSA / VC-DRSA parameter value for which given boundary has been calculatedboundaryAttributesMask
- mask for attributes allowing boundary calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute has not been taken into account
at all when calculating boundary. If attributesMask(i) is different than zero,
than corresponding attribute has been taken into account when calculating boundary.InvalidValueException
- when VC-IRSA / VC-DRSA parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getInconsistentExamplesForExtendedLowerApproximation(double variableConsistencyParameterValue)
Integer
) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter value (variable consistency approach).
For example in case of upward union of decision classes, inconsistent examples are those, which dominate
examples from extended lower approximation but do not belong to the considered union.
In case of downward union of decision classes, inconsistent examples are those, which are dominated by
examples from extended lower approximation but do not belong to the considered union.
For single decision class, inconsistent examples are those, which are indiscernible with examples from the extended lower approximation
but they do not belong to the considered single decision class.variableConsistencyParameterValue
- variable consistency parameter value for which extended lower approximation should be calculated (VC-IRSA / VC-DRSA)Integer
) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter valueInvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic java.util.HashSet<java.lang.Integer> getInconsistentExamplesForExtendedLowerApproximation(double variableConsistencyParameterValue, byte[] attributesMask)
Integer
) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter value (variable consistency approach)
and for given mask for attributes.
For example in case of upward union of decision classes, inconsistent examples are those, which dominate
examples from extended lower approximation but do not belong to the considered union.
In case of downward union of decision classes, inconsistent examples are those, which are dominated by
examples from extended lower approximation but do not belong to the considered union.
For single decision class, inconsistent examples are those, which are indiscernible with examples from the extended lower approximation
but they do not belong to the considered single decision class.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which extended lower approximation should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing extended lower approximation calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating lower approximation. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating extended lower approximation.Integer
) numbers of negative examples which are inconsistent with (among) examples from
extended lower approximation of this entity, calculated for given variable consistency parameter value
and for given mask for attributesInvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getPOSRegionExamples()
Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA). This method should return examples belonging to
lower approximation of this entity.Integer
) numbers of examples belonging to positive (POS) region of this entitypublic java.util.HashSet<java.lang.Integer> getPOSRegionExamples(byte[] attributesMask)
Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes.
This method returns examples belonging to lower approximation of this entity,
calculated for the most restrictive variable consistency parameter value and for given mask for attributes.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.attributesMask
- Mask for attributes allowing positive region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating positive region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating positive region.Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for given mask for attributesInvalidValueException
- when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getPOSRegionExamples(double variableConsistencyParameterValue)
Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA). Returned set is a sum of two sets.variableConsistencyParameterValue
- variable consistency parameter value for which POS region should be calculated (VC-IRSA / VC-DRSA)Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter valueInvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic java.util.HashSet<java.lang.Integer> getPOSRegionExamples(double variableConsistencyParameterValue, byte[] attributesMask)
Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes.
Returned set is a sum of two sets.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which POS region should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing positive region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating positive region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating positive region.Integer
) numbers of examples belonging to positive (POS) region of this entity,
calculated for given variable consistency parameter value and for given mask for attributesInvalidValueException
- when given variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getNEGRegionExamples()
Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA).
In case of IRSA, this method returns examples belonging to POS region of the complement of decision class.
In case of DRSA, this method returns examples belonging to POS region of the opposite union
(e.g., for union "at least 3" opposite union is "at most 2").getNEGRegionExamples(byte[])
and passes null
as mask for attributes to indicate
that mask is not used.Integer
) numbers of examples belonging to negative (NEG) region of this entitypublic abstract java.util.HashSet<java.lang.Integer> getNEGRegionExamples(byte[] attributesMask)
Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes.
In case of IRSA, this method should return examples belonging to POS region of the complement of decision class, calculated for given mask.
In case of DRSA, this method should return examples belonging to POS region of the opposite union, calculated for given mask
(e.g. for union "at least 3" opposite union is "at most 2").null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.attributesMask
- Mask for attributes allowing negative region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating negative region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating negative region.Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given mask for attributesInvalidValueException
- when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getNEGRegionExamples(double variableConsistencyParameterValue)
Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA).
In case of VC-IRSA, when POS regions of different decision classes may overlap,
NEG region of the considered decision class Cl is a set difference between POS region of the complement of Cl
and POS region of Cl.
In case of VC-DRSA, when POS regions may overlap, NEG region consists of those examples belonging to
POS region of the opposite union of decision classes (e.g., for union "at least 3", opposite union is "at most 2"),
which are not in POS region of the considered union at the same time.getNEGRegionExamples(double, byte[])
and passes null
as mask for attributes to indicate
that mask is not used.variableConsistencyParameterValue
- variable consistency parameter value for which NEG region should be calculated (VC-IRSA / VC-DRSA)Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter valueInvalidValueException
- when given variable consistency parameter value is outside [0, 1] rangepublic abstract java.util.HashSet<java.lang.Integer> getNEGRegionExamples(double variableConsistencyParameterValue, byte[] attributesMask)
Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes.
In case of VC-IRSA, when POS regions of different decision classes may overlap,
NEG region of the considered decision class Cl is a set difference between POS region of the complement of Cl
and POS region of Cl.
In case of VC-DRSA, when POS regions may overlap, NEG region consists of those examples belonging to
POS region of the opposite union of decision classes (e.g., for union "at least 3", opposite union is "at most 2"),
which are not in POS region of the considered union at the same time.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which NEG region should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing negative region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating negative region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating negative region.Integer
) numbers of examples belonging to negative (NEG) region of this entity,
calculated for given variable consistency parameter value and for given mask for attributesInvalidValueException
- when variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getBNDRegionExamples()
Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA).
In case of IRSA and DRSA, this method returns examples belonging to boundary of this entity.getBNDRegionExamples(byte[])
and passes null
as mask for attributes to indicate,
that mask for attributes is not used.Integer
) numbers of examples belonging to boundary (BND) region of this entitypublic java.util.HashSet<java.lang.Integer> getBNDRegionExamples(byte[] attributesMask)
Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for the most restrictive variable consistency parameter value (IRSA / DRSA) and for given mask for attributes.
In case of IRSA and DRSA, this method returns examples belonging to boundary
of this entity, defined for given mask for attributes.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.attributesMask
- Mask for attributes allowing boundary region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating boundary region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating boundary region.Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given mask for attributesInvalidValueException
- when length of the mask for attributes is different than quantity of attributes in memory containerpublic java.util.HashSet<java.lang.Integer> getBNDRegionExamples(double variableConsistencyParameterValue)
Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA).
Returns the complement of POS + NEG regions of considered entity.getBNDRegionExamples(double, byte[])
and passes null
as mask for attributes to indicate,
that mask for attributes is not used.variableConsistencyParameterValue
- variable consistency parameter value for which BND region should be calculated (VC-IRSA / VC-DRSA)Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter valueInvalidValueException
- when variable consistency parameter value is outside [0, 1] rangepublic java.util.HashSet<java.lang.Integer> getBNDRegionExamples(double variableConsistencyParameterValue, byte[] attributesMask)
Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter value (VC-IRSA / VC-DRSA) and for given mask for attributes.
Returns the complement of POS + NEG regions of considered entity.null
.
If attributesMask(i) = 0 then corresponding attribute should be ignored.
Any non-zero value means that corresponding attribute should be considered.
In fact only mask values corresponding to active and condition attributes are important.
All other values are not taken into account and may be any.variableConsistencyParameterValue
- variable consistency parameter value for which BND region should be calculated (VC-IRSA / VC-DRSA)attributesMask
- Mask for attributes allowing boundary region calculations for a subset
of active and condition attributes.
It is checked if length of the mask is equal to the quantity of attributes in memory container.
If attributesMask(i) is equal to zero, than corresponding attribute is not taken into account
at all when calculating boundary region. If attributesMask(i) is different than zero,
than corresponding attribute is taken into account when calculating boundary region.Integer
) numbers of examples belonging to boundary (BND) region of this entity,
calculated for given variable consistency parameter value and for given mask for attributesInvalidValueException
- when variable consistency parameter value is outside [0, 1] range
or when length of the mask for attributes is different than quantity of attributes in memory containerpublic abstract boolean variableConsistencyParameterValueIsTheMostRestrictiveOne(double variableConsistencyParameterValue)
variableConsistencyParameterValue
- variable consistency parameter value to be checkedtrue
if given variable consistency parameter value is the most restrictive one, false
otherwisepublic abstract double getTheMostRestrictiveVariableConsistencyParameterValue()
public abstract java.lang.String getDescription()
public abstract int getDecisionAttributeNumber()
public java.lang.Object clone()
clone
in class java.lang.Object