k -mean alignment for curve clustering

The problem of curve clustering when curves are misaligned is considered. A novel algorithm is described, which jointly clusters and aligns curves. The proposed procedure efficiently decouples amplitude and phase variability; in particular, it is able to detect amplitude clusters while simultaneousl...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computational statistics & data analysis Ročník 54; číslo 5; s. 1219 - 1233
Hlavní autoři: Sangalli, Laura M., Secchi, Piercesare, Vantini, Simone, Vitelli, Valeria
Médium: Journal Article
Jazyk:angličtina
Vydáno: Amsterdam Elsevier B.V 01.05.2010
Elsevier
Edice:Computational Statistics & Data Analysis
Témata:
[formula omitted]-mean algorithm
Curve alignment
Curve clustering
Exact sciences and technology
Functional analysis
Functional data analysis
Functional data analysis Curve alignment Curve clustering k-mean algorithm
General topics
Mathematical analysis
Mathematics
Multivariate analysis
Numerical analysis
Numerical analysis. Scientific computation
Numerical methods in probability and statistics
Probability and statistics
Sciences and techniques of general use
Statistics
k -mean algorithm
Curve clustering
Functional data analysis
Curve alignment
Discriminant analysis
Data analysis
Cluster analysis (statistics)
Functional analysis
k-mean algorithm
Multivariate analysis
Statistical simulation
Algorithm
Mean estimation
Statistical computation
ISSN:0167-9473, 1872-7352
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The problem of curve clustering when curves are misaligned is considered. A novel algorithm is described, which jointly clusters and aligns curves. The proposed procedure efficiently decouples amplitude and phase variability; in particular, it is able to detect amplitude clusters while simultaneously disclosing clustering structures in the phase, pointing out features that can neither be captured by simple curve clustering nor by simple curve alignment. The procedure is illustrated via simulation studies and applications to real data.
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2009.12.008