Constraint Programming for Data Mining

In this talk I shall explore the relationship between constraint-based mining and constraint programming. In particular, I shall show how the typical constraints used in pattern mining can be formulated for use in constraint programming environments. The resulting framework is surprisingly flexible and allows one to combine a wide range of mining constraints in different ways. The approach is implemented in off-the-shelf constraint programming systems and evaluated empirically. The results show that the approach is not only very expressive, but also works well on complex benchmark problems. In addition to providing a detailed account of our actual initial results for item-set mining, I shall also argue that the use of constraint programming techniques and methodologies provides a new and interesting paradigm for data mining. The work I will report on is joint work with Tias Guns and Siegfried Nijssen.