4.7 Article

A new shape-based clustering algorithm for time series

Journal

INFORMATION SCIENCES
Volume 609, Issue -, Pages 411-428

Publisher

ELSEVIER SCIENCE INC
DOI: 10.1016/j.ins.2022.07.105

Keywords

Time series clustering; Fractional correlation; Shape shift deviating; Time series clustering; Fractional correlation; Shape shift deviating

Funding

  1. National Natural Science Foundation of China [62172082, 62072084,62072086]
  2. Fundamental Research Funds for the central Universities [N2116008]

Ask authors/readers for more resources

This paper discusses the research on time series clustering and the importance of shape-based clustering algorithms. It proposes a new algorithm called FrOKShape, which shows excellent results when combined with traditional clustering algorithms and comparable performance to existing algorithms.
Time series clustering is a research hotspot in data mining. Most of the existing clustering algorithms combine with the classical distance measure which ignore the offset of sequence shape. As a result, shape-based clustering algorithms are becoming increasingly popular. On the majority of data sets, the most representative shape-based clustering algo-rithm, KShape, which defines a shape-based distance with shift invariance, has been shown to outperform other algorithms.In this paper, we propose a new shape-based clustering algorithm named Fractional Order Shape-based k-cluster(FrOKShape), which defines a multi-variable shape-based dis-tance by normalized fractional order cross-correlation and uses the DTW Barycenter Averaging (DBA) as a center computation strategy. Our distance exhibits excellent shape shift deviating properties and good compatibility integrated with a variety of existing clus-tering center strategies so that it can provide more potential good results. Experiments show that combining our distance with a traditional clustering algorithm produces excel-lent clustering indicators. In a series of comparative experiments, FrOKShape also exhibits a comparable result to the existing better shape-based clustering algorithm KShape.(c) 2022 Elsevier Inc. All rights reserved.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available