... for everything else, there's publication databases

The quick round-up

I am currently a post-doctoral researcher at the Computer Science department of the Catholic University Leuven (Katholieke Universiteit Leuven), in the Machine Learning group which is in turn a subgroup of the Declarative Languages and Artificial Intelligence (DTAI) group.

My PhD thesis was on "Mining Sets of Patterns", a topic that grew out of replacing heuristic local pattern mining components, which made me realize that many existing techniques for classification, feature construction, and clustering can be considered to consist of such a local pattern step as well as a pattern set mining step (either as post-processing or as outer loop).

This realization gives a powerful toolbox in which lots of techniques can be mixed and matched, yet with many approaches comes much responsibility...to properly evaluate these approaches. This is a task that has arguably been somewhat ignored in the pattern mining literature so far and I am currently very interested in all its aspects: how to generate data that enables such comparisons, how to establish the similarity of different data sets, and of course the actual evaluation itself and what we can learn about the effectiveness of established methods.


Short CV - status December 2012