InformationTable: R6 class representing an information table.
In jaspeir/NIJ_Tabitha: Dominance-based Rough Set Approach and the DOMLEM Algorithm
Description
Details
Public fields
Methods
Description
An information table consists of the decision table and meta-data.
Details
This class stores the decision table, and meta-data.
The decision table consists of an object identifier column, a decision column, and at least one additional attribute.
The meta-data consist of the attribute names, their types, and the alpha and beta values for similarity attributes.
Public fields
decisionTablethe set of examples
metaDatameta-data of the attributes, including their name and type, along with alpha and beta parameters for similarity variables
objectsvector of object names
Methods
Public methods
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Method new()
Create a new information table object.
Usage
InformationTable$new(decisionTable, metaData = NA)
Arguments
decisionTabledata frame containing the decision examples
metaDatadata frame containing the meta-data of the attributes. This parameter is optional, and if not provided, we assume all dominance attributes.
Method isCompatible()
Method to determine whether another information table is compatible with this one.
Usage
InformationTable$isCompatible(it)
Arguments
itthe information table to compare to
Method getType()
Method to get the type of an attribute.
Usage
InformationTable$getType(attribute)
Arguments
attributethe name of the attribute
Returns
the type
Method removeObjects()
Method for creating a new information table by removing examples of belonging to the specified objects.
Usage
InformationTable$removeObjects(objects)
Arguments
objectsthe object to filter out
Returns
a new information table instance
Method downwardClassUnion()
Method for calculating the downward class union.
Usage
InformationTable$downwardClassUnion(class)
Arguments
classthe decision class to compare to
Returns
the set of objects in the downward class union
Method upwardClassUnion()
Method for calculating the upward class union.
Usage
InformationTable$upwardClassUnion(class)
Arguments
classthe decision class to compare to
Returns
the set of objects in the upward class union
Method encodeDecisionColumn()
Method for encoding the decision column to the 1:N range, where N is the cardinality of this column.
Usage
InformationTable$encodeDecisionColumn()
Returns
the encoded decision column
Method decodeDecisions()
Method for decoding 1:N-encoded decisions.
Usage
InformationTable$decodeDecisions(encoded)
Arguments
encodeda vector of encoded decisions
Returns
a vector with the decoded decisions
Method classUnions()
Method for calculating all downward- and upward class unions at once.
Usage
InformationTable$classUnions()
Returns
a pair of matrices for both class unions, where each row represents a class, and each column represents an object
Method partitionAttributes()
Method that partitions attribute set P into into sets of the same attribute type.
Only types relevant for the dominance relation are considered (indiscernibility, similarity, and dominance).
Usage
InformationTable$partitionAttributes(P)
Arguments
Pthe set of attributes to partition - vector of attribute names
Returns
a list of attribute sets
Function to determine whether x dominates y on the mixed attribute set P.
Method dominates()
Usage
InformationTable$dominates(x, y, P, compareSimilaritySwitched = FALSE)
Arguments
xthe left operand - object name
ythe right operand - object name
Pthe set of attributes to test - vector of attribute names
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
whether x dominates y on attribute set P
Method dominatingAndDominatedSets()
Method for calculating the P-dominated and P-dominating sets all at once.
Usage
InformationTable$dominatingAndDominatedSets(P)
Arguments
Pthe set of attributes to test - vector of attribute names
Method to determine whether x is similar to y on attribute q.
Method similar()
Usage
InformationTable$similar(x, y, q)
Arguments
xthe left operand - a data frame
ythe right operand - a data frame
qthe attribute to test
Returns
whether x is similar to y on attribute q
Method dominatingSet()
This method calculates the dominating set of an object with respect to a criterion set.
Usage
InformationTable$dominatingSet(x, P, compareSimilaritySwitched = TRUE)
Arguments
xthe object - object name
Pthe criterion set
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
the set of objects that dominate object x
Method dominatedSet()
This method calculates the dominated set of an object with respect to a criterion set.
Usage
InformationTable$dominatedSet(x, P, compareSimilaritySwitched = FALSE)
Arguments
xthe object - object name
Pthe criterion set
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
the set of objects that are dominated by object x
Method upwardClassUnionUpperApproximation()
This method calculates the P-upper approximations of the upward class unions.
Usage
InformationTable$upwardClassUnionUpperApproximation(dominating_U)
Arguments
dominating_Uthe P-dominating sets (U) - matrix
Returns
the approximations for all classes in a boolean matrix from
Method upwardClassUnionLowerApproximation()
This method calculates the P-lower approximations of the upward class unions.
Usage
InformationTable$upwardClassUnionLowerApproximation(downward_U)
Arguments
downward_Uthe P-upper approximations of the downward class unions - matrix
Returns
the approximations for all classes in a boolean matrix from
Method downwardClassUnionUpperApproximation()
This method calculates the P-upper approximations of the downward class unions.
Usage
InformationTable$downwardClassUnionUpperApproximation(dominated_U)
Arguments
dominated_Uthe P-dominated sets (U) - matrix
Returns
the approximations for all classes in a boolean matrix from
Method downwardClassUnionLowerApproximation()
This method calculates the P-lower approximations of the downward class unions.
Usage
InformationTable$downwardClassUnionLowerApproximation(upward_U)
Arguments
upward_Uthe P-upper approximations of the upward class unions - matrix
Returns
the approximations for all classes in a boolean matrix from
Method roughSets()
This method calculates the P-lower and P-upper approximations of class unions and boundary regions.
Usage
InformationTable$roughSets(P)
Arguments
Pthe attribute set
Returns
a named list of the approximations
Method boundaryRegions()
This method calculates the boundary regions of rough sets.
Usage
InformationTable$boundaryRegions(roughSets)
Arguments
roughSetsthe class union approximations.
Returns
the upward and downward boundary regions
Method accuracyOfApproximation()
This method calculates the accuracy of the approximations of the class unions.
Usage
InformationTable$accuracyOfApproximation(roughSets)
Arguments
roughSetsthe approximations
Returns
a pair of vectors describing the accuracy of the downward and upward class union approximations
Method qualityOfApproximation()
This method calculates the quality of the approximations of the class unions.
Usage
InformationTable$qualityOfApproximation(boundaryRegions)
Arguments
boundaryRegionsthe boundary regions of the rough set
Returns
a number expressing the ratio of all P-correctly sorted actions to all actions in the decision table
Method clone()
The objects of this class are cloneable with this method.
Usage
InformationTable$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
jaspeir/NIJ_Tabitha documentation built on June 3, 2020, 12:38 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Details Public fields Methods
Description
An information table consists of the decision table and meta-data.
Details
This class stores the decision table, and meta-data. The decision table consists of an object identifier column, a decision column, and at least one additional attribute. The meta-data consist of the attribute names, their types, and the alpha and beta values for similarity attributes.
Public fields
decisionTablethe set of examples
metaDatameta-data of the attributes, including their name and type, along with alpha and beta parameters for similarity variables
objectsvector of object names
Methods
Public methods
Method new()
Create a new information table object.
Usage
InformationTable$new(decisionTable, metaData = NA)
Arguments
decisionTabledata frame containing the decision examples
metaDatadata frame containing the meta-data of the attributes. This parameter is optional, and if not provided, we assume all dominance attributes.
Method isCompatible()
Method to determine whether another information table is compatible with this one.
Usage
InformationTable$isCompatible(it)
Arguments
itthe information table to compare to
Method getType()
Method to get the type of an attribute.
Usage
InformationTable$getType(attribute)
Arguments
attributethe name of the attribute
Returns
the type
Method removeObjects()
Method for creating a new information table by removing examples of belonging to the specified objects.
Usage
InformationTable$removeObjects(objects)
Arguments
objectsthe object to filter out
Returns
a new information table instance
Method downwardClassUnion()
Method for calculating the downward class union.
Usage
InformationTable$downwardClassUnion(class)
Arguments
classthe decision class to compare to
Returns
the set of objects in the downward class union
Method upwardClassUnion()
Method for calculating the upward class union.
Usage
InformationTable$upwardClassUnion(class)
Arguments
classthe decision class to compare to
Returns
the set of objects in the upward class union
Method encodeDecisionColumn()
Method for encoding the decision column to the 1:N range, where N is the cardinality of this column.
Usage
InformationTable$encodeDecisionColumn()
Returns
the encoded decision column
Method decodeDecisions()
Method for decoding 1:N-encoded decisions.
Usage
InformationTable$decodeDecisions(encoded)
Arguments
encodeda vector of encoded decisions
Returns
a vector with the decoded decisions
Method classUnions()
Method for calculating all downward- and upward class unions at once.
Usage
InformationTable$classUnions()
Returns
a pair of matrices for both class unions, where each row represents a class, and each column represents an object
Method partitionAttributes()
Method that partitions attribute set P into into sets of the same attribute type. Only types relevant for the dominance relation are considered (indiscernibility, similarity, and dominance).
Usage
InformationTable$partitionAttributes(P)
Arguments
Pthe set of attributes to partition - vector of attribute names
Returns
a list of attribute sets Function to determine whether x dominates y on the mixed attribute set P.
Method dominates()
Usage
InformationTable$dominates(x, y, P, compareSimilaritySwitched = FALSE)
Arguments
xthe left operand - object name
ythe right operand - object name
Pthe set of attributes to test - vector of attribute names
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
whether x dominates y on attribute set P
Method dominatingAndDominatedSets()
Method for calculating the P-dominated and P-dominating sets all at once.
Usage
InformationTable$dominatingAndDominatedSets(P)
Arguments
Pthe set of attributes to test - vector of attribute names Method to determine whether x is similar to y on attribute q.
Method similar()
Usage
InformationTable$similar(x, y, q)
Arguments
xthe left operand - a data frame
ythe right operand - a data frame
qthe attribute to test
Returns
whether x is similar to y on attribute q
Method dominatingSet()
This method calculates the dominating set of an object with respect to a criterion set.
Usage
InformationTable$dominatingSet(x, P, compareSimilaritySwitched = TRUE)
Arguments
xthe object - object name
Pthe criterion set
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
the set of objects that dominate object x
Method dominatedSet()
This method calculates the dominated set of an object with respect to a criterion set.
Usage
InformationTable$dominatedSet(x, P, compareSimilaritySwitched = FALSE)
Arguments
xthe object - object name
Pthe criterion set
compareSimilaritySwitchedwhether to test similarity with the parameters switched
Returns
the set of objects that are dominated by object x
Method upwardClassUnionUpperApproximation()
This method calculates the P-upper approximations of the upward class unions.
Usage
InformationTable$upwardClassUnionUpperApproximation(dominating_U)
Arguments
dominating_Uthe P-dominating sets (U) - matrix
Returns
the approximations for all classes in a boolean matrix from
Method upwardClassUnionLowerApproximation()
This method calculates the P-lower approximations of the upward class unions.
Usage
InformationTable$upwardClassUnionLowerApproximation(downward_U)
Arguments
downward_Uthe P-upper approximations of the downward class unions - matrix
Returns
the approximations for all classes in a boolean matrix from
Method downwardClassUnionUpperApproximation()
This method calculates the P-upper approximations of the downward class unions.
Usage
InformationTable$downwardClassUnionUpperApproximation(dominated_U)
Arguments
dominated_Uthe P-dominated sets (U) - matrix
Returns
the approximations for all classes in a boolean matrix from
Method downwardClassUnionLowerApproximation()
This method calculates the P-lower approximations of the downward class unions.
Usage
InformationTable$downwardClassUnionLowerApproximation(upward_U)
Arguments
upward_Uthe P-upper approximations of the upward class unions - matrix
Returns
the approximations for all classes in a boolean matrix from
Method roughSets()
This method calculates the P-lower and P-upper approximations of class unions and boundary regions.
Usage
InformationTable$roughSets(P)
Arguments
Pthe attribute set
Returns
a named list of the approximations
Method boundaryRegions()
This method calculates the boundary regions of rough sets.
Usage
InformationTable$boundaryRegions(roughSets)
Arguments
roughSetsthe class union approximations.
Returns
the upward and downward boundary regions
Method accuracyOfApproximation()
This method calculates the accuracy of the approximations of the class unions.
Usage
InformationTable$accuracyOfApproximation(roughSets)
Arguments
roughSetsthe approximations
Returns
a pair of vectors describing the accuracy of the downward and upward class union approximations
Method qualityOfApproximation()
This method calculates the quality of the approximations of the class unions.
Usage
InformationTable$qualityOfApproximation(boundaryRegions)
Arguments
boundaryRegionsthe boundary regions of the rough set
Returns
a number expressing the ratio of all P-correctly sorted actions to all actions in the decision table
Method clone()
The objects of this class are cloneable with this method.
Usage
InformationTable$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.