Ensembles of random sphere cover classifiers

We propose and evaluate a new set of ensemble methods for the Randomised Sphere Cover (RSC) classifier. RSC is a classifier using the sphere cover method that bases classification on distance to spheres rather than distance to instances. The randomised nature of RSC makes it ideal for use in ensembles. We propose two ensemble methods tailored to the RSC classifier; RSE, an ensemble based on instance resampling and RSSE, a subspace ensemble. We compare RSE and RSSE to tree based ensembles on a set of UCI datasets and demonstrates that RSC ensembles perform significantly better than some of these ensembles, and not significantly worse than the others. We demonstrate via a case study on six gene expression data sets that RSSE can outperform other subspace ensemble methods on high dimensional data when used in conjunction with an attribute filter. Finally, we perform a set of Bias/Variance decomposition experiments to analyse the source of improvement in comparison to a base classifier.