RoughSets (version 1.0-0)

RI.indiscernibilityBasedRules.RST: Rule induction based on RST

Description

It is a function generating rule induction based on indiscernibility classes.

Usage

RI.indiscernibilityBasedRules.RST(decision.table,
    feature.set)

Arguments

decision.table

a "DecisionTable" class representing the decision table. See SF.asDecisionTable.

feature.set

a "FeatureSubset" class which is produced by feature selection functions based on RST e.g. FS.permutation.heuristic.reduct.RST, etc.

See

Value

• A class "RuleSetRST" containing the following components/attributes:
  • rules: every rule which is a list consisting of five parts as follows:
    • idx: it is a vector of indexes of attributes involved in each rule.
    • values: it is a vector of values of each attributes used as the antecedent part.
    • consequent: it is a vector of a value of decision attribute as the consequent part.
    • support: it is a numeric value representing the support value of each rule.
    • laplace: it is a numeric value representing the laplace value of the consequent part.
  • uniqueCls: it is a vector containing the decision classes.
  • clsProbs: it is vector showing the probability of the decision classes.
  • majorityCls: it is a value representing the majority value in the rules.
  • method: it is the type of used method. In this case, it is"indiscernibilityBasedRules".

Details

Basically, this function uses the output of feature selection step as input data. So, before calling this function, we execute a particular function implementing feature selection,

e.g. FS.permutation.heuristic.reduct.RST.

It should be noted that this function only allows classification problems. After obtaining the rules, predicting can be done by calling predict or predict.RuleSetRST.

See Also

predict.RuleSetFRST, RI.GFRS.FRST, and RI.hybridFS.FRST.

Examples

###########################################################
## Example
##############################################################
data(RoughSetData)
decision.table <- RoughSetData$hiring.dt

## determine feature subset/reduct
reduct <- FS.permutation.heuristic.reduct.RST(decision.table,  permutation = NULL)

rules <- RI.indiscernibilityBasedRules.RST(decision.table, reduct)