Time series feature learning with labeled and unlabeled data

Wang, H; Zhang, Q; Wu, J; Pan, S; Chen, Y

Time series feature learning with labeled and unlabeled data

Wang, H Zhang, Q Wu, J

Pan, S

Chen, Y

Permalink

Publication Type:: Journal Article
Citation:: Pattern Recognition, 2019, 89 pp. 55 - 66
Issue Date:: 2019-05-01

Closed Access

	Filename	Description	Size
	1-s2.0-S0031320318304473-main.pdf	Published Version	2.02 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wang, H	en_US
dc.contributor.author	Zhang, Q	en_US
dc.contributor.author	Wu, J https://orcid.org/0000-0002-1371-5801	en_US
dc.contributor.author	Pan, S https://orcid.org/0000-0003-0794-527X	en_US
dc.contributor.author	Chen, Y	en_US
dc.date.issued	2019-05-01	en_US
dc.identifier.citation	Pattern Recognition, 2019, 89 pp. 55 - 66	en_US
dc.identifier.issn	0031-3203	en_US
dc.identifier.uri	http://hdl.handle.net/10453/138539
dc.description.abstract	© 2018 Elsevier Ltd Time series classification has attracted much attention in the last two decades. However, in many real-world applications, the acquisition of sufficient amounts of labeled training data is costly, while unlabeled data is usually easily to be obtained. In this paper, we study the problem of learning discriminative features (segments) from both labeled and unlabeled time series data. The discriminative segments are often referred to as shapelets. We present a new Semi-Supervised Shapelets Learning (SSSL for short) model to efficiently learn shapelets by using both labeled and unlabeled time series data. Briefly, SSSL engages both labeled and unlabeled time series data in an integrated model that considers the least squares regression, the power of the pseudo-labels, shapelets regularization, and spectral analysis. The experimental results on real-world data demonstrate the superiority of our approach over existing methods.	en_US
dc.relation.ispartof	Pattern Recognition	en_US
dc.relation.isbasedon	10.1016/j.patcog.2018.12.026	en_US
dc.subject.classification	Artificial Intelligence & Image Processing	en_US
dc.title	Time series feature learning with labeled and unlabeled data	en_US
dc.type	Journal Article
utslib.citation.volume	89	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	0806 Information Systems	en_US
utslib.for	0906 Electrical and Electronic Engineering	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - CAI - Centre for Artificial Intelligence
pubs.organisational-group	/University of Technology Sydney/Students
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US
pubs.volume	89	en_US

Abstract:

© 2018 Elsevier Ltd Time series classification has attracted much attention in the last two decades. However, in many real-world applications, the acquisition of sufficient amounts of labeled training data is costly, while unlabeled data is usually easily to be obtained. In this paper, we study the problem of learning discriminative features (segments) from both labeled and unlabeled time series data. The discriminative segments are often referred to as shapelets. We present a new Semi-Supervised Shapelets Learning (SSSL for short) model to efficiently learn shapelets by using both labeled and unlabeled time series data. Briefly, SSSL engages both labeled and unlabeled time series data in an integrated model that considers the least squares regression, the power of the pseudo-labels, shapelets regularization, and spectral analysis. The experimental results on real-world data demonstrate the superiority of our approach over existing methods.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/138539