Finding Closed Frequent Itemsets in Linear Time

Efficient algorithms for mining frequent itemsets are crucial for the analysis of data arising in a broad of applications. Recent studies proved that mining frequent closed itemsets substantially reduces the number of itemsets to be presented meanwhile has the same power as frequent itemsets mining. This paper presents a very simple and efficient algorithm, CFII, for mining frequent closed itemsets. This algorithm is based on an alternative, but equivalent representation of the target problem. Theoretically, the new algorithm could complete in linear time according to the number of frequent closed item sets.