Prefix-Projection Global Constraint for Sequential Pattern Mining

摘要

Sequential pattern mining under constraints is a challenging data miningtask. Many efficient ad hoc methods have been developed for mining sequentialpatterns, but they are all suffering from a lack of genericity. Recent workshave investigated Constraint Programming (CP) methods, but they are not stilleffective because of their encoding. In this paper, we propose a globalconstraint based on the projected databases principle which remedies to thisdrawback. Experiments show that our approach clearly outperforms CP approachesand competes well with ad hoc methods on large datasets.