Local projections for high-dimensional outlier detection | TU Wien

Information

Publication Type: Journal Paper (without talk)
Workgroup(s)/Project(s): not specified
Date: ongoing
DOI: 10.1007/s40300-020-00183-5

Abstract

A novel approach for outlier detection is proposed, called local projections, which is based on concepts of the Local Outlier Factor (LOF) (Breunig et al. in Lof: identifying densitybased local outliers. In: ACM sigmod record, ACM, volume 29, pp. 93-104, 2000) and ROBPCA (Hubert et al. in Technometrics 47(1):64-79, 2005). By using aspects of both methods, this algorithm is robust towards noise variables and is capable of performing outlier detection in multi-group situations. The idea is to focus on local descriptions of the observations and their neighbors using linear projections. The outlyingness of an observation is determined by a weighted distance of the observation to all identified projection spaces, with weights depending on the appropriateness of the local description. Experiments with simulated and real data demonstrate the usefulness of this method when compared to existing outlier detection algorithms.

Additional Files and Images

No additional files or images.

Weblinks

BibTeX

@article{Ortner-2021,
  title =      "Local projections for high-dimensional outlier detection",
  author =     "Thomas Ortner and Peter Filzmoser and Maia Rohm and Sarka
               Brodinova and Christian Breiteneder",
  abstract =   "A novel approach for outlier detection is proposed, called
               local projections, which is based on concepts of the Local
               Outlier Factor (LOF) (Breunig et al. in Lof: identifying
               densitybased local outliers. In: ACM sigmod record, ACM,
               volume 29, pp. 93-104, 2000) and ROBPCA (Hubert et al. in
               Technometrics 47(1):64-79, 2005). By using aspects of both
               methods, this algorithm is robust towards noise variables
               and is capable of performing outlier detection in
               multi-group situations. The idea is to focus on local
               descriptions of the observations and their neighbors using
               linear projections. The outlyingness of an observation is
               determined by a weighted distance of the observation to all
               identified projection spaces, with weights depending on the
               appropriateness of the local description. Experiments with
               simulated and real data demonstrate the usefulness of this
               method when compared to existing outlier detection
               algorithms.",
  doi =        "10.1007/s40300-020-00183-5",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/ongoing/Ortner-2021/",
}