CANCELED due to illness This meeting will be in our regular room at the regular time: SGG B1.09 at 14:00 Topic:  Comparing predictive model performance on reconviction data Preparation: Read at least the abstract of Nikolaj’s paper Content: On finding the best predictive model on three data sets – on general, violent, and sexual recidivism –…

 Boosting: fun and delicious. Erik-Jan explains boosting while gesticulating wildly The group launches into a tense competition for the ultimate error rate Maarten is the cunning champion of the spiral dataset. Bon appetit! Emmeke easily levels her competition in the second round This meeting was at 14:00 in room B1.09 Topic: Boosting as a way of…

This meeting was on Thursday at 14:00 in room A3.17. During this meeting, Kees introduced Support Vector Machines, a famous “black-box” classification and regression method. For those who want to try out SVM with the Caret R package, here you can find a nice practical introduction. Certainly the kernel trick will be part of the discussion. A great blog…

This meeting was on the 27th at 14:00 in room A3.17! We discussed the same paper, starting with a presentation by Emmeke to refresh our memory on how regression trees and random forests work. Then we went into how these were adapted to longitudinal models by the authors in the RE-EM framework. Matthieu Brinkhuis (UD Computer Science)…

Edit this excerpt before posting! [click to read]