Abstract
In data mining approach, the quantitative attributes should be appropriately dealt with as well as the Boolean attributes. This paper presents an essential improvement for extracting fuzzy association rules from database. In this paper, we define equivalence of fuzzy itemsets and related theorems as a new concept for fuzzy data mining. Then, we propose a basic algorithm based on the Apriori algorithm for rule extraction utilizing equivalence of the fuzzy itemsets based on redundancy concepts of fuzzy association rules. The essential performance of the algorithm is evaluated through numerical experiments using benchmark data. From the results, the method is found to be promising in terms of computational time and redundant rule pruning.
