Chapter 2 already illustrated why only an empirical approach to algorithm selection is likely to succeed in the foreseeable future. Now that monitoring mechanisms are in place to feed observations into a dedicated performance database, as described in chapter 5, it has to be discussed how this data can be analyzed so that suitable selection mappings are generated (see fig. 6.1).


Feature Selection Performance Data Algorithm Selection Selection Tree Selector Generation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2012

Authors and Affiliations

  • Roland Ewald

There are no affiliations available

Personalised recommendations