A Fast Algorithm for Mining Closed Itemsets under the Length-Decreasing Support Constraint

Mining frequent itemsets or patterns is a fundamental and essential problem in many data mining application. Because of the inherent computational complexity, mining the complete set of frequent patterns remains to be a difficult task. Mining closed patterns is a good solution to the problem. And previous study has show that mining frequent patterns with length-decreasing support constraint is very helpful in removing some uninteresting patterns. Therefore, in this paper, we study how to mine closed itemsets under length-decreasing support constraint. We have proposed several new pruning methods to enhance the closed itemset mining under new constraint, and developed an efficient algorithm, LDS_CLOSED. Experimental results show that LDS_CLOSED not only generates more concise result set, but also runs much faster than the existing mining algorithm, DCI_CLOSED