Practical approach to outlier detection using support vector regression

  • Authors:

  • Junya Nishiguchi;Chosei Kaseda;Hirotaka Nakayama;Masao Arakawa;Yeboon Yun

  • Affiliations:

  • Yamatake Corporation;Yamatake Corporation;Konan University;Kagawa University;Kagawa University

  • Venue:
  • ICONIP'08 Proceedings of the 15th international conference on Advances in neuro-information processing - Volume Part I
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

For precise estimation with soft sensors, it is necessary to remove outliers from the measured raw data before constructing the model. Conventionally, visualization and maximum residual error have been used for outlier detection, but they often fail to detect outliers for nonlinear function with multidimensional input. In this paper we propose a practical approach to outlier detection using Support Vector Regression, which reduces computational cost and defines outlier threshold appropriately. We apply this approach to both test and industrial datasets for validation.