Dedicated Shapelet Distance Engine for Time-Series Classification

This article proposes a parameterizable parallel hardware accelerator for addressing the drastic computing requirements of current shapelet-based methods.

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE design and test 2022-04, Vol.39 (2), p.7-14
Hauptverfasser:	Costa, Victor Oliveira, de Araujo Gewehr, Carlos Gabriel, Vicenzi, Julio Costella, Carara, Everton Alceu, de Oliveira, Leonardo Londero
Format:	Magazinearticle
Sprache:	eng
Schlagworte:	Euclidean distance Hardware acceleration hardware accelerator Parallel processing Ports (computers) shapelets Time complexity Time series analysis time-series classification
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	This article proposes a parameterizable parallel hardware accelerator for addressing the drastic computing requirements of current shapelet-based methods.
ISSN:	2168-2356 2168-2364
DOI:	10.1109/MDAT.2021.3102884