DistanceBased Outlier Detection: Consolidation and Renewed Bearing

Free registration required

Executive Summary

Detecting outliers in data is an important problem with interesting applications in a myriad of domains ranging from data cleaning to financial fraud detection and from network intrusion detection to clinical diagnosis of diseases. Over the last decade of research, distance-based outlier detection algorithms have emerged as a viable, scalable, parameter-free alternative to the more traditional statistical approaches. In this paper the authors assess several distance-based outlier detection approaches and evaluate them.

  • Format: PDF
  • Size: 665.1 KB