USING DIFFSET FOR MINING FREQUENT WEIGHTED CLOSED ITEMSETS ON WEIGHTED ITEMS TRANSACTION DATABASES

Abstract

Mining frequent itemsets plays an important role in mining association rules. For weighted items transaction databases, mining frequent weighted closed itemsets (FWCIs) is one of the method proposed by author. However, for dense databases, the mining time of FWCIs is still high. In this paper, an algorithm for mining frequent weighted closed itemsets using diffset (FWCIs-DIFF) is proposed. Some theorems are presented first, base on them, an algorithm for mining FWCIs-DIFF is proposed. For dense databases, experimental results show that the mining time of FWCIs-DIFF is always smaller than that of FWCIs.
Keyword: mining frequent itemsets, mining frequent weighted closed itemset, diffset.