2013
DOI: 10.1007/s10618-013-0322-1
|View full text |Cite
|
Sign up to set email alerts
|

Classification of time series by shapelet transformation

Abstract: Time-series classification (TSC) problems present a specific challenge for classification algorithms: how to measure similarity between series. A \emph{shapelet} is a time-series subsequence that allows for TSC based on local, phase-independent similarity in shape. Shapelet-based classification uses the similarity between a shapelet and a series as a discriminatory feature. One benefit of the shapelet approach is that shapelets are comprehensible, and can offer insight into the problem domain. The original sha… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

2
271
0
29

Year Published

2017
2017
2024
2024

Publication Types

Select...
3
2
2

Relationship

1
6

Authors

Journals

citations
Cited by 459 publications
(323 citation statements)
references
References 33 publications
2
271
0
29
Order By: Relevance
“…Otolith boundaries are also extracted and represented, or encoded in different ways (transformed) prior to analysis with methods such as Fourier transforms (Begg and Brown 2000;Galley et al 2006;Bani et al 2013); and Elliptical Fourier transforms (Campana and Casselman 1993;Duarte-Neto et al 2008). Other methods of otolith boundary representation include Wavelets (Parisi-Baradad et al 2005), Curvature-Scale-Space (Begg et al 2005;Parisi-Baradad et al 2005) and the more recent Shapelet transform method (Lines et al 2012;Mapp et al 2013;Hills et al 2014).…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Otolith boundaries are also extracted and represented, or encoded in different ways (transformed) prior to analysis with methods such as Fourier transforms (Begg and Brown 2000;Galley et al 2006;Bani et al 2013); and Elliptical Fourier transforms (Campana and Casselman 1993;Duarte-Neto et al 2008). Other methods of otolith boundary representation include Wavelets (Parisi-Baradad et al 2005), Curvature-Scale-Space (Begg et al 2005;Parisi-Baradad et al 2005) and the more recent Shapelet transform method (Lines et al 2012;Mapp et al 2013;Hills et al 2014).…”
Section: Introductionmentioning
confidence: 99%
“…The Waikato Environment for Knowledge Analysis, (WEKA (Hall et al 2009)) is a freely available library of machine learning tools which are widely adopted by the machine learning community, and which have previously been used for otolith classification (Hills et al 2014). The library comprises many statistical and modelling tools together with learning algorithms.…”
Section: Introductionmentioning
confidence: 99%
“…Também não há experimentos extensivos que avaliam as medidas de qualidade de subsequências no domínio da transformada. E o principal problema de técnicas que envolvem subsequências permanece sem solução satisfatória, pois a quantidade de subsequências possíveis de um conjunto de dados é enorme, e a avaliação de cada uma delas é um processo custoso, logo em seus trabalhos eles realizaram experimentos em subsequências de tamanho restrito (HILLS et al, 2014).…”
Section: Justificativa E Motivaçãounclassified
“…A primeira das lacunas está no fato de que os experimentos apresentados até o momento foram em um conjunto reduzido do total de shapelets por questões de tempo de execução. Por exemplo, Hills et al (2014) propôs um algoritmo que restringe a busca por bons shapelets àqueles que possuam uma quantidade de amostras entre um valor mínimo e máximo. A nossa crítica é que sem experimentos que utilizem todos os shapelets qualquer avaliação, seja de acurácia ou de tempo de execução, de métodos que reduzem o espaço de busca dos shapelets tem baixo valor científico.…”
Section: Lacunas and Objetivosunclassified
See 1 more Smart Citation