We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Novel semi-metrics for multivariate change point analysis and anomaly detection

Abstract: This paper proposes a new method for determining similarity and anomalies between time series, most practically effective in large collections of (likely related) time series, by measuring distances between structural breaks within such a collection. We introduce a class of \emph{semi-metric} distance measures, which we term \emph{MJ distances}. These semi-metrics provide an advantage over existing options such as the Hausdorff and Wasserstein metrics. We prove they have desirable properties, including better sensitivity to outliers, while experiments on simulated data demonstrate that they uncover similarity within collections of time series more effectively. Semi-metrics carry a potential disadvantage: without the triangle inequality, they may not satisfy a "transitivity property of closeness." We analyse this failure with proof and introduce an computational method to investigate, in which we demonstrate that our semi-metrics violate transitivity infrequently and mildly. Finally, we apply our methods to cryptocurrency and measles data, introducing a judicious application of eigenvalue analysis.
Comments: Accepted manuscript. Minor edits since v2. Equal contribution from first two authors
Subjects: Machine Learning (cs.LG); Dynamical Systems (math.DS); Computation (stat.CO); Methodology (stat.ME); Machine Learning (stat.ML)
Journal reference: Physica D: Nonlinear Phenomena 412 (2020) 132636
DOI: 10.1016/j.physd.2020.132636
Cite as: arXiv:1911.00995 [cs.LG]
  (or arXiv:1911.00995v3 [cs.LG] for this version)

Submission history

From: Nicholas James [view email]
[v1] Mon, 4 Nov 2019 00:04:30 GMT (13749kb,D)
[v2] Fri, 7 Feb 2020 14:58:46 GMT (6630kb,D)
[v3] Fri, 3 Jul 2020 10:44:42 GMT (5550kb,D)

Link back to: arXiv, form interface, contact.