Parallel Techniques for Variable Size Segmentation of Time Series Datasets - LIRMM - Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier
Conference Papers Year : 2022

Parallel Techniques for Variable Size Segmentation of Time Series Datasets

Lamia Djebour
  • Function : Author
  • PersonId : 1119503
Reza Akbarinia
Florent Masseglia

Abstract

Given the high data volumes in time series applications, or simply the need for fast response times, it is usually necessary to rely on alternative, shorter representations of these series, usually with loss. This incurs approximate comparisons of time series where precision is a major issue.In this paper, we propose a new parallel approach for segmenting time series before their transformation into symbolic representations. It can reduce significantly the error incurred by possible splittings at different steps of the representation calculation, by taking into account the sum of squared errors (SSE). This is particularly useful for time series similarity search, which is the core of many data analytics tasks. We provide theoretical guarantees on the lower bound of similarity measures, and our experiments illustrate that our technique can improve significantly the time series representation quality.
Fichier principal
Vignette du fichier
SAE-ADBIS.pdf (511.18 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

lirmm-03805997 , version 1 (07-10-2022)

Identifiers

Cite

Lamia Djebour, Reza Akbarinia, Florent Masseglia. Parallel Techniques for Variable Size Segmentation of Time Series Datasets. ADBIS 2022 - 26th European Conference on Advances in Databases and Information Systems, Sep 2022, Turin, Italy. pp.148-162, ⟨10.1007/978-3-031-15740-0_12⟩. ⟨lirmm-03805997⟩
47 View
85 Download

Altmetric

Share

More