Fuzzy clustering of time series in the frequency domain

Elizabeth Maharaj, Pierpaolo D'Urso

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Traditional and fuzzy cluster analyses are applicable to variables whose values are uncorrelated. Hence, in order to cluster time series data which are usually serially correlated, one needs to extract features from the time series, the values of which are uncorrelated. The periodogram which is an estimator of the spectral density function of a time series is a feature that can be used in the cluster analysis of time series because its ordinates are uncorrelated. Additionally, the normalized periodogram and the logarithm of the normalized periodogram are also features that can be used. In this paper, we consider a fuzzy clustering approach for time series based on the estimated cepstrum. The cepstrum is the spectrum of the logarithm of the spectral density function. We show in our simulation studies for the typical generating processes that have been considered, fuzzy clustering based on the cepstral coefficients performs very well compared to when it is based on other features.
Original languageEnglish
Pages (from-to)1187 - 1211
Number of pages25
JournalInformation Sciences
Volume181
Issue number7
DOIs
Publication statusPublished - 2011

Cite this

Maharaj, Elizabeth ; D'Urso, Pierpaolo. / Fuzzy clustering of time series in the frequency domain. In: Information Sciences. 2011 ; Vol. 181, No. 7. pp. 1187 - 1211.
@article{cc2c8b0716574a37a27d194280c677aa,
title = "Fuzzy clustering of time series in the frequency domain",
abstract = "Traditional and fuzzy cluster analyses are applicable to variables whose values are uncorrelated. Hence, in order to cluster time series data which are usually serially correlated, one needs to extract features from the time series, the values of which are uncorrelated. The periodogram which is an estimator of the spectral density function of a time series is a feature that can be used in the cluster analysis of time series because its ordinates are uncorrelated. Additionally, the normalized periodogram and the logarithm of the normalized periodogram are also features that can be used. In this paper, we consider a fuzzy clustering approach for time series based on the estimated cepstrum. The cepstrum is the spectrum of the logarithm of the spectral density function. We show in our simulation studies for the typical generating processes that have been considered, fuzzy clustering based on the cepstral coefficients performs very well compared to when it is based on other features.",
author = "Elizabeth Maharaj and Pierpaolo D'Urso",
year = "2011",
doi = "10.1016/j.ins.2010.11.031",
language = "English",
volume = "181",
pages = "1187 -- 1211",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier",
number = "7",

}

Fuzzy clustering of time series in the frequency domain. / Maharaj, Elizabeth; D'Urso, Pierpaolo.

In: Information Sciences, Vol. 181, No. 7, 2011, p. 1187 - 1211.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Fuzzy clustering of time series in the frequency domain

AU - Maharaj, Elizabeth

AU - D'Urso, Pierpaolo

PY - 2011

Y1 - 2011

N2 - Traditional and fuzzy cluster analyses are applicable to variables whose values are uncorrelated. Hence, in order to cluster time series data which are usually serially correlated, one needs to extract features from the time series, the values of which are uncorrelated. The periodogram which is an estimator of the spectral density function of a time series is a feature that can be used in the cluster analysis of time series because its ordinates are uncorrelated. Additionally, the normalized periodogram and the logarithm of the normalized periodogram are also features that can be used. In this paper, we consider a fuzzy clustering approach for time series based on the estimated cepstrum. The cepstrum is the spectrum of the logarithm of the spectral density function. We show in our simulation studies for the typical generating processes that have been considered, fuzzy clustering based on the cepstral coefficients performs very well compared to when it is based on other features.

AB - Traditional and fuzzy cluster analyses are applicable to variables whose values are uncorrelated. Hence, in order to cluster time series data which are usually serially correlated, one needs to extract features from the time series, the values of which are uncorrelated. The periodogram which is an estimator of the spectral density function of a time series is a feature that can be used in the cluster analysis of time series because its ordinates are uncorrelated. Additionally, the normalized periodogram and the logarithm of the normalized periodogram are also features that can be used. In this paper, we consider a fuzzy clustering approach for time series based on the estimated cepstrum. The cepstrum is the spectrum of the logarithm of the spectral density function. We show in our simulation studies for the typical generating processes that have been considered, fuzzy clustering based on the cepstral coefficients performs very well compared to when it is based on other features.

U2 - 10.1016/j.ins.2010.11.031

DO - 10.1016/j.ins.2010.11.031

M3 - Article

VL - 181

SP - 1187

EP - 1211

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

IS - 7

ER -