Machine Learning
Support vector domain description
Pattern Recognition Letters - Special issue on pattern recognition in practice VI
Kernel Whitening for One-Class Classification
SVM '02 Proceedings of the First International Workshop on Pattern Recognition with Support Vector Machines
Combining One-Class Classifiers
MCS '01 Proceedings of the Second International Workshop on Multiple Classifier Systems
Kernel Fisher Discriminants for Outlier Detection
Neural Computation
Estimating the Support of a High-Dimensional Distribution
Neural Computation
Using an Ensemble of One-Class SVM Classifiers to Harden Payload-based Anomaly Detection Systems
ICDM '06 Proceedings of the Sixth International Conference on Data Mining
Kernel PCA for novelty detection
Pattern Recognition
Soft clustering using weighted one-class support vector machines
Pattern Recognition
One-Class Classification by Combining Density and Class Probability Estimation
ECML PKDD '08 Proceedings of the 2008 European Conference on Machine Learning and Knowledge Discovery in Databases - Part I
Bagging One-Class Decision Trees
FSKD '08 Proceedings of the 2008 Fifth International Conference on Fuzzy Systems and Knowledge Discovery - Volume 02
Hi-index | 0.00 |
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.