International Workshop on Machine Learning and Data Mining in Pattern Recognition

MLDM 2007: Machine Learning and Data Mining in Pattern Recognition pp 61-75

Outlier Detection with Kernel Density Functions

  • Longin Jan Latecki
  • Aleksandar Lazarevic
  • Dragoljub Pokrajac
Conference paper

DOI: 10.1007/978-3-540-73499-4_6

Volume 4571 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Latecki L.J., Lazarevic A., Pokrajac D. (2007) Outlier Detection with Kernel Density Functions. In: Perner P. (eds) Machine Learning and Data Mining in Pattern Recognition. MLDM 2007. Lecture Notes in Computer Science, vol 4571. Springer, Berlin, Heidelberg

Abstract

Outlier detection has recently become an important problem in many industrial and financial applications. In this paper, a novel unsupervised algorithm for outlier detection with a solid statistical foundation is proposed. First we modify a nonparametric density estimate with a variable kernel to yield a robust local density estimation. Outliers are then detected by comparing the local density of each point to the local density of its neighbors. Our experiments performed on several simulated data sets have demonstrated that the proposed approach can outperform two widely used outlier detection algorithms (LOF and LOCI).

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Longin Jan Latecki
    • 1
  • Aleksandar Lazarevic
    • 2
  • Dragoljub Pokrajac
    • 3
  1. 1.CIS Dept. Temple University Philadelphia, PA 19122USA
  2. 2.United Technology Research Center 411 Silver Lane, MS 129-15 East Hartford, CT 06108USA
  3. 3.CIS Dept. CREOSA and AMRC, Delaware State University, Dover DE 19901USA