Amir Ebrahimzadeh - University, Mashhad branch
Reza Sheibani - Department of Computer, Mashhad Branch, Islamic Azad University Mashhad, iran

Mining association rules is one of the important research problems in data mining. So, many algorithms have been proposed to find association rules in databases with either binary or quantitative attributes. One of these approaches is fuzzy association rules mining. Fuzzy Apriori and its different variations are the only popular fuzzy association rule mining (ARM) algorithms available today. Like the crisp version of Apriori, fuzzy Apriori is a very slow and inefficient algorithm for very large datasets. Hence ,in this paper, we present An efficient algorithm named FCT. This method discovers fuzzy association rules by scaning the database once, and performing three tasks simultaneously .First , compute the fuzzy supports of candidate 1-itemsets and then generate large 1-itemsets.Second , divide database into multiple cluster tables ,such that transaction with length k , fall into cluster table k.Third ,builds new structure called CDi , for each cluster table i, such that CDi[A,x]=ΣμA(x) ,where x is an item and A denotes linguistic term. Then fuzzy large itemsets are generated according to the cluster tables, instead of scanning whole the database. In addition , if CDi[A,x]=0 for cluster i and item i, then for computing the fuzzy support of each candidate itemset containing A(x) , scanning this cluster can be ignored.Consequently , we reduce incredible amount of scanning data and therefore the running time of mining algorithm is reduced greatly. Experimental results show the efficiency of the presented approach for real world transactions.

Fuzzy Association Rulse;, cluster table