Common Dissimilarity Measures are Inappropriate for Time Series Clustering

Authors

  • Cássio Martini Martins Pereira Instituto de Ciências Matemáticas e de Computação - ICMC - USP
  • Rodrigo F. de Mello Universidade de São Paulo

DOI:

https://doi.org/10.22456/2175-2745.25070

Abstract

Clustering algorithms have been actively used to identify similar timeseries, providing a better understanding of data. However, common clustering dis-similarity measures disregard time series correlations, yielding poor results. In thispaper, we introduce a dissimilarity measure based on series partial autocorrelations.Experiments compare hierarchical clustering algorithms using the common dissimi-larity measures, such as Euclidean Distance and Dynamic Time Warping, to clustertime series following Box-Jenkins Auto-Regressive models. Results show that ourdissimilarity measure produces better results for both synthetic and real data sets interms of the Adjusted Rand Index and Normalized Hubert Γ statistic. Our findingsconfirm that the choice of dissimilarity measure is crucial for improving time seriesclustering quality.

Downloads

Download data is not yet available.

Published

2013-01-09

How to Cite

Pereira, C. M. M., & Mello, R. F. de. (2013). Common Dissimilarity Measures are Inappropriate for Time Series Clustering. Revista De Informática Teórica E Aplicada, 20(1), 25–48. https://doi.org/10.22456/2175-2745.25070

Issue

Section

Regular Papers