Mining Frequent Weighted Closed Itemsets

摘要

Mining frequent itemsets plays an important role in mining association rules. One of methods for mining frequent itemsets is mining frequent weighted itemsets (FWIs). However, the number of FWIs is often very large when the database is large. Besides, FWIs will generate a lot of rules and some of them are redundant. In this paper, a method for mining frequent weighted closed itemsets (FWCIs) in weighted items transaction databases is proposed. Some theorems are derived first, and based on them, an algorithm for mining FWCIs is proposed. Experimental results show that the number of FWCIs is always smaller than that of FWIs and the mining time is also better.