Pushing Tougher Constraints in Frequent Pattern Mining

In this paper we extend the state-of-art of the constraints that can be pushed in a frequent pattern computation. We introduce a new class of tough constraints, namely Loose Anti-monotone constraints, and we deeply characterize them by showing that they are a superclass of convertible anti-monotone constraints (e.g. constraints on average or median) and that they model tougher constraints (e.g. constraints on variance or standard deviation). Then we show how these constraints can be exploited in a level-wise Apriori-like computation by means of a new data-reduction technique: the resulting algorithm outperforms previous proposals for convertible constraints, and it is to treat much tougher constraints with the same effectiveness of easier ones.