A graph-based algorithm for frequent closed itemsets mining

Frequent itemsets mining plays an essential role in data mining, but it often generates a large number of redundant itemsets that reduce the efficiency of the mining task. Frequent closed itemsets are subset of frequent itemsets, but they contain all information of frequent itemsets. The most existing methods of frequent closed itemset mining are apriori-based. The efficiency of those methods is limited to the repeated database scan and the candidate set generation. We propose a graph-based algorithm for mining frequent closed itemsets called GFCG (graph-based frequent closed itemset generation). The new algorithm constructs an association graph to represent the frequent relationship between items, and recursively generates frequent closed itemset based on that graph. It scans the database for only two times, and avoids candidate set generation. GFCG outperforms a priori-based algorithm in experiment study and shows good performance both in speed and scale up properties.