3 papers
stat.ME2017
Robust and sparse k-means clustering for high-dimensional data
Sarka Brodinova, Peter Filzmoser, Thomas Ortner +2
In real-world application scenarios, the identification of groups poses a significant challenge due to possibly occurring outliers and existing noise variables. Therefore, there is…
stat.ME2017
Multigroup discrimination based on weighted local projections
Thomas Ortner, Irene Hoffmann, Peter Filzmoser +3
A novel approach for supervised classification analysis for high dimensional and flat data (more variables than observations) is proposed. We use the information of class-membershi…
stat.ME2017
Local projections for high-dimensional outlier detection
Thomas Ortner, Peter Filzmoser, Maia Zaharieva +2
In this paper, we propose a novel approach for outlier detection, called local projections, which is based on concepts of Local Outlier Factor (LOF) (Breunig et al., 2000) and RobP…