15 January 2020 in IEEE Transactions on Cybernetics
Time-series clustering is the process of grouping time series with respect to their similarity or characteristics. Previous approaches usually combine a specific distance measure for time series and a standard clustering method. However, these approaches do not take the similarity of the different subsequences of each time series into account, which can be used to better compare the time-series objects of the dataset. In this article, we propose a novel technique of time-series clustering consisting of two clustering stages. In a first step, a least-squares polynomial segmentation procedure is applied to each time series, which is based on a growing window technique that returns different-length segments. Then, all of the segments are projected into the same dimensional space, based on the coefficients of the model that approximates the segment and a set of statistical features. After mapping, a first hierarchical clustering phase is applied to all mapped segments, returning groups of segments for each time series. These clusters are used to represent all time series in the same dimensional space, after defining another specific mapping process. In a second and final clustering stage, all the time-series objects are grouped. We consider internal clustering quality to automatically adjust the main parameter of the algorithm, which is an error threshold for the segmentation. The results obtained on 84 datasets from the UCR Time Series Classification Archive have been compared against three state-of-the-art methods, showing that the performance of this methodology is very promising, especially on larger datasets.
1 May 2016 in Ocean Engineering
AMA.C. Martínez-Estudillo
1 September 2018 in Applied Soft Computing
26 November 2014 in Climate Dynamics
This paper proposes a time series segmentation algorithm combining a clustering technique and a genetic algorithm to automatically find segments sharing common statistical characteristics in paleoclimate time series. The segments are transformed into a six-dimensional space composed of six statistical measures, most of which have been previously considered in the detection of warning signals of critical transitions. Experimental results show that the proposed approach applied to paleoclimate data could effectively analyse Dansgaard–Oeschger (DO) events and uncover commonalities and differences in their statistical and possibly their dynamical characterisation. In particular, warning signals were robustly detected in the GISP2 and NGRIP \(\delta ^{18}\hbox {O}\) ice core data for several DO events (e.g. DO 1, 4, 8 and 12) in the form of an order of magnitude increase in variance, autocorrelation and mean square distance from a linear approximation (i.e. the mean square error). The increase in mean square error, suggesting nonlinear behaviour, has been found to correspond with an increase in variance prior to several DO events for \(\sim \) 90 % of the algorithm runs for the GISP2 \(\delta ^{18}\hbox {O}\) dataset and for \(\sim \) 100 % of the algorithm runs for the NGRIP \(\delta ^{18}\hbox {O}\) dataset. The proposed approach applied to well-known dynamical systems and paleoclimate datasets provides a novel visualisation tool in the field of climate time series analysis.
1 February 2018 in Applied Soft Computing
1 August 2019 in Neurocomputing