Tutorial: Preference-based Pattern Mining

Preference-based Pattern Mining – Half-day tutorial (Tuesday morning) 


Speakers: All speakers have extensive experience in research on pattern mining and how to discover useful patterns.

Bruno Crémilleux – Université de Caen (France)
He received his PhD in computer science in 1991 from the University of Grenoble. He is professor in computer science since 2005 at the University of Caen-Normandy. His main research interests are pattern (set) discovery, Constraint Satisfaction Problems and data mining, preference queries and exploratory data mining. He was co-chair of several ECML/PKDD workshops.

Marc Plantevit – Université de Lyon (France)
He received his PhD in computer science in 2008 from the University of Montpellier. He has been an associate professor in the computer science department of the University of Lyon since 2009. His research interest include constraint-based pattern mining in general. Currently, he is very interested with sophisticate pattern domains (dynamic/ attributed graphs) and in incorporating background knowledge into pattern mining.

Arnaud Soulet – Université François Rabelais de Tours (France)
He received his PhD in 2006 from the University of Caen. He is currently associate professor in computer science since 2007 at the University François Rabelais of Tours. He has an expertise in constraint-based pattern mining and involvement in the mining process like pattern mining techniques for preference elicitation.


Tutorial description: This tutorial focuses on the recent shift from constraint-based pattern mining to preference-based pattern mining and interactive pattern mining. Constraint-based pattern mining, which shares common notions with FCA, is now a mature domain of data mining that makes it possible to handle various different pattern domains (e.g., itemsets, sequences, graphs) with a large variety of constraints thanks to solid theoretical foundations and an efficient algorithmic machinery. Even though, it has been realized for a long time that it is difficult for the end-user to model her interest in term of constraints and above to overcome the well-known thresholding issue, researchers have only recently intensified their study of methods for finding high-quality patterns according to the user’s preferences.  In this tutorial, we discuss the need of preferences in pattern mining, the principles and methods of the use of preferences in pattern mining. Many methods are derived from constraint-based pattern mining by integrating utility functions or interestingness measures as quantitative preference model. This approach transforms pattern mining in an optimization problem guided by user specified preferences. However, in practice, the user has only a vague idea of what useful patterns could be. The recent research field of interactive pattern mining relies on the automatic acquisition of these preferences and the development of the instant data mining field.



Comments are closed