Ensemble Learning with Evolutionary Computation: Application to Feature Ranking
Kees Jong, Elena Marchiori and Michele Sebag
In: PPSN, Parallel Problem Solving from Nature, sept. 2004, Birmingham.
A new ensemble approach for Feature Selection is presented, aggregating the feature rankings extracted from the hypotheses extracted along independent runs with evolutionary learning.
A statistical model is devised to enable the direct evaluation
of the approach; comparative experimental results show its good behavior on non-linear concepts when the features outnumber the examples.
|Postscript - PASCAL Members only - Requires a viewer, such as GhostView|