Towards improving solution dominance with incomparability conditions:a case-study using Generator Itemset Mining

Finding interesting patterns is a challenging task in data mining. Constraint based mining is a well-known approach to this, and one for which constraint programming has been shown to be a well-suited and generic framework.Dominance programming has been proposed as an extension that can capture aneven wider class of constraint-based mining problems, by allowing to comparerelations between patterns. In this paper, in addition to specifying a dominancerelation, we introduce the ability to specify an incomparability condition. Usingthese two concepts we devise a generic framework that can do a batch-wise searchthat avoids checking incomparable solutions. We extend the ESSENCE languageand underlying modelling pipeline to support this. We use generator itemset mining problem as a test case and give a declarative specification for that. We alsopresent preliminary experimental results on this specific problem class with a CPsolver backend to show that using the incomparability condition during searchcan improve the efficiency of dominance programming and reduces the need forpost-processing to filter dominated solutions