Stacking for Ensembles of Local Experts in Metabonomic Applications
- Cite this paper as:
- Lienemann K., Plötz T., Fink G.A. (2009) Stacking for Ensembles of Local Experts in Metabonomic Applications. In: Benediktsson J.A., Kittler J., Roli F. (eds) Multiple Classifier Systems. MCS 2009. Lecture Notes in Computer Science, vol 5519. Springer, Berlin, Heidelberg
Recently, Ensembles of local experts have successfully been applied for the automatic detection of drug-induced organ toxicities based on spectroscopic data. For suitable Ensemble composition an expert selection optimization procedure is required that identifies the most relevant classifiers to be integrated. However, it has been observed that Ensemble optimization tends to overfit on the training data. To tackle this problem we propose to integrate a stacked classifier optimized via cross-validation that is based on the outputs of local experts. In order to achieve probabilistic outputs of Support Vector Machines used as local experts we apply a sigmoidal fitting approach. The results of an experimental evaluation on a challenging data set from safety pharmacology demonstrate the improved generalizability of the proposed approach.
Unable to display preview. Download preview PDF.