Mining frequent itemsets with tough constraints

In order to efficiently sift through a large number of mined rules, constraint-based mining is introduced. Two large classes of constraints -monotone constraints and succinct constraints have been investigated. However, the problem of frequent itemsets mining with tough constraints has not been solved because of the complexity of the constraints. In this paper, we propose two methods which use the order as the pre-process to solve this problem. The first method is to push the tough constraints deeply inside the candidate generation-and-test approach such as Apriori. The second is to combine the constraints with pattern-growth methods such as FP-tree.