Detection of anomalous time series based on multiple distance measures

Huaming Huang, Kishan G. Mehrotra, Chilukuri K. Mohan

Research output: Chapter in Book/Entry/PoemConference contribution

1 Scopus citations

Abstract

Automatic detection of anomalous series is an important task and several approaches have been suggested using a single detection measure. We propose a multi-measure based approach and compare it with existing methods. To select a short list of effective measures we perform extensive evaluations for several combinations of proposed measures. Our results show that the proposed algorithm is able to detect a variety of anomalies in datasets from different domains. Our approach outperforms existing methods that use single measures, which detect only some types of anomalies.

Original languageEnglish (US)
Title of host publication28th International Conference on Computers and Their Applications 2013, CATA 2013
Pages147-152
Number of pages6
StatePublished - 2013
Event28th International Conference on Computers and Their Applications 2013, CATA 2013 - Honolulu, HI, United States
Duration: Mar 4 2013Mar 6 2013

Publication series

Name28th International Conference on Computers and Their Applications 2013, CATA 2013

Other

Other28th International Conference on Computers and Their Applications 2013, CATA 2013
Country/TerritoryUnited States
CityHonolulu, HI
Period3/4/133/6/13

Keywords

  • Anomalous time series detection
  • Distance measures
  • Outlier detection
  • Rank-based methods
  • Time series

ASJC Scopus subject areas

  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Detection of anomalous time series based on multiple distance measures'. Together they form a unique fingerprint.

Cite this