Constraint-based mining using constraint solving

ICTEAM + MLG seminar

Event Speaker: Tias Guns
Event Place: Room Paul Otlet (Reaumur Building)
Event Date: 10/20/2015 - 12:0

The use of constraints is prevalent in data mining, most often to express background knowledge or feedback from the user. Computationally, constraints have been studied for many years by the constraint solving community. In recent years, such constraint technology is increasingly used in the field of constraint-based mining too. Many challenges exist though, at the level of modelling the problem (encodings, high-level languages, new primitives) as well as at the solving level (scalability, redundant constraints, search strategies). We review recurring motivations and highlight some of the challenges with MiningZinc, a high-level pattern mining language with solving mechanism that can hybridize specialised algorithms and constraint solvers.