Document RAIRO Oper. Res. 42 (2008) 69-83
DOI: 10.1051/ro:2008004
Combining evolutionary algorithms and exact approaches for multi-objective knowledge discovery
Mohammed Khabzaoui, Clarisse Dhaenens and El-Ghazali TalbiPolytech'Lille, LIFL - CNRS / INRIA - Bâtiment M3, University of Lille 1, 59655 Villeneuve d'Ascq Cedex, France; Clarisse.Dhaenens@lifl.fr; El-Ghazali.Talbi@lifl.fr
(Received October 30, 2006. Accepted October 30, 2007. Published online 21 February 2008.)
Abstract
An important task of knowledge discovery deals with discovering association rules. This very general model has been widely studied and efficient algorithms have been proposed. But most of the time, only frequent rules are seeked. Here we propose to consider this problem as a multi-objective combinatorial optimization problem in order to be able to also find non frequent but interesting rules. As the search space may be very large, a discussion about different approaches is proposed and a hybrid approach that combines a metaheuristic and an exact operator is presented.
Mathematics Subject Classification. 90Cxx, 68XX
Key words: Hybridization, multi-objective optimization, knowledge discovery, association rules
© EDP Sciences, ROADEF, SMAI 2008