Mining fuzzy association rules

Abstract

A technique called F-APACS which employs linguistic terms to represent the revealed regularities and exceptions for mining fuzzy association rules is presented. The linguistic representation is useful when the rules discovered are presented to human experts for examination. The definition of the linguistic terms is based on fuzzy set theory and the rules having these terms are called fuzzy association rules. The use of fuzzy techniques is considered as one of the key components of data mining systems because of the affinity with the human knowledge representation.