Exportar Publicação

Cardoso, M. G. M. S., Martins, A. A. A. F. & Lagarto, J. (2022). The clustering performance of a weighted combined distance between time series. 17th conference of the International Federation of Classification Societies .

M. M. Cardoso et al.,  "The clustering performance of a weighted combined distance between time series", in 17th conference of the Int. Federation of Classification Societies , Porto, 2022

@misc{cardoso2022_1776098004339,
	author = "Cardoso, M. G. M. S. and Martins, A. A. A. F. and Lagarto, J.",
	title = "The clustering performance of a weighted combined distance between time series",
	year = "2022",
	url = "https://ifcs2022.fep.up.pt/"
}

TY  - CPAPER
TI  - The clustering performance of a weighted combined distance between time series
T2  - 17th conference of the International Federation of Classification Societies 
AU  - Cardoso, M. G. M. S.
AU  - Martins, A. A. A. F.
AU  - Lagarto, J.
PY  - 2022
CY  - Porto
UR  - https://ifcs2022.fep.up.pt/
AB  - Recently, [1],we proposed a newdissimilarity measure between time series - COMB,
a uniform convex combination of four (normalized) distance measures: Euclidean;
Pearson correlation based; Periodogram based; and a distance between estimated autocorrelation
structures. In this work, we propose a method to determine the weights
of the convex combination of distances in COMB: it relies on the concordance
of clusterings obtained by each individual distance measure and COMB derived
clustering. A weighted COMB measure is thus obtained, WCOMB. We then test
the clustering performance of WCOMB vs. COMB by conducting an experimental
analysis on all the time series datasets of the UCR archive. We evaluate the concordance
between the clusters obtained using K-Medoids and the original classes (using
adjusted Rand index) as well as the cohesion-separation of the clusters (using the
Silhouette index). In addition, we consider a clustering application - with data from
the Portuguese Transmission System Operator, on time series of electricity consumption
(2014 to 2019) - to compare the performance of both methods. Significant
differences between the average Silhouette values of clusters obtained were found.
The concordance with the original classes’ structure exhibits similar performance in
both approaches. We conclude that, for unsupervised leaning, it can be worthwhile
to invest on deriving specific weights for the distances integrating COMB.
ER  -