ABBA: adaptive Brownian bridge-based symbolic aggregation of time series

DOI10.1007/S10618-020-00689-6MaRDI QIDQ2194039zbMATH OpenOpenAlexFDO

Authors Steven Elsworth, Stefan Güttel

Publication date 25 August 2020

Published in Data Mining and Knowledge Discovery (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2003.12469

zbMATH Keywords

Brownian bridgedimension reductionsymbolic aggregationtime series

Mathematics Subject Classification ID

Software, source code, etc. for problems pertaining to statistics (62-04) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Abstract: A new symbolic representation of time series, called ABBA, is introduced. It is based on an adaptive polygonal chain approximation of the time series into a sequence of tuples, followed by a mean-based clustering to obtain the symbolic representation. We show that the reconstruction error of this representation can be modelled as a random walk with pinned start and end points, a so-called Brownian bridge. This insight allows us to make ABBA essentially parameter-free, except for the approximation tolerance which must be chosen. Extensive comparisons with the SAX and 1d-SAX representations are included in the form of performance profiles, showing that ABBA is able to better preserve the essential shape information of time series compared to other approaches. Advantages and applications of ABBA are discussed, including its in-built differencing property and use for anomaly detection, and Python implementations provided.

Recommendations

• A non-parametric symbolic approximate representation for long time series
• Similarity measure based on numerical symbolic and shape feature for time series
• Fast and Accurate Time-Series Clustering
• An approach to dimensionality reduction in time series
• Scalable time series classification

Cites work

• A review on distance based time series classification
• A Space-Economical Suffix Tree Construction Algorithm
• Benchmarking optimization software with performance profiles.
• Detection of change in persistence of a linear time series
• scientific article; zbMATH DE number 5506203 (Why is no real title available?)
• Learning a symbolic representation for multivariate time series classification
• MODL: a Bayes optimal discretization method for continuous attributes
• Scikit-learn: machine learning in Python

Cited in

(2)

• A non-parametric symbolic approximate representation for long time series
• A new symbolic representation method for time series

Describes a project that uses

Uses Software

• MODL
• iSAX
• Scikit

This page was built for publication: ABBA: adaptive Brownian bridge-based symbolic aggregation of time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2194039)