Distance-based feature selection on classification of uncertain objects

Abstract

We study the problem of classification on uncertain objects whose locations are uncertain and described by probability density functions (pdf). We propose a novel supervised UK-means algorithm for classifying uncertain objects to overcome the computation bottleneck of existing algorithms. Additionally, we consider to select features that can capture the relevant properties of uncertain data. We experimentally demonstrate that our proposed approaches are more efficient than existing algorithms and can attain comparatively accurate results on non-overlapping data sets. © 2011 Springer-Verlag.