The main advantage of Constraint Programming (CP) approaches for sequential pattern mining (SPM) is their modularity, which includes the ability to add new constraints (regular expressions, length restrictions, etc).
However, the existing CP-based approaches are not as scalable as some of the most advanced mining systems. We show how, using techniques from both data mining and CP, one can use a generic constraint solver and yet outperform both CP-based and specialized approaches. The data and software related to these works are available here
John is a 2nd year Ph.D. student at Université Catholique de Louvain. His research topics are the hybridization of CP and Data mining, especially the design of new constraints for sequential pattern mining. He is working on how to handle constraints over datasets with timestamps.