Time series classification maps time series to labels. The nearest neighbour algorithm (NN) using the Dynamic Time Warping (DTW) similarity measure is a leading algorithm for this task. NN compares each time series to be classified to every time series in the training database. With a training database of N time series of lengths L, each classification requires ϑ(N · L 2 ) computations. The databases used in almost all prior research have been relatively small (with less than 10, 000 samples) and much of the research has focused on making DTW's complexity linear with L, leading to a runtime complexity of O(N · L). As we demonstrate with an example in remote sensing, real-world time series databases are now reaching the million-to-billion scale. This wealth of training data brings the promise of higher accuracy, but raises a significant challenge because N is becoming the limiting factor. As DTW is not a metric, indexing objects induced by its space is extremely challenging. We tackle this task in this paper. We develop TSI, a novel algorithm for Time Series Indexing which combines a hierarchy of K-means clustering with DTW-based lower-bounding. We show that, on large databases, TSI makes it possible to classify time series orders of magnitude faster than the state of the art.

No abstract

We propose MultiRocket, a fast time series classification (TSC) algorithm that achieves state-of-the-art accuracy with a tiny fraction of the time and without the complex ensembling structure of many state-of-the-art methods. MultiRocket improves on MiniRocket, one of the fastest TSC algorithms to date, by adding multiple pooling operators and transformations to improve the diversity of the features generated. In addition to processing the raw input series, MultiRocket also applies first order differences to transform the original series. Convolutions are applied to both representations, and four pooling operators are applied to the convolution outputs. When benchmarked using the University of California Riverside TSC benchmark datasets, MultiRocket is significantly more accurate than MiniRocket, and competitive with the best ranked current method in terms of accuracy, HIVE-COTE 2.0, while being orders of magnitude faster.

Time series classification maps time series to labels. The nearest neighbor algorithm (NN) using the Dynamic Time Warping (DTW) similarity measure is a leading algorithm for this task and a component of the current best ensemble classifiers for time series. However, NN-DTW is only a winning combination when its meta-parameter -its warping window -is learned from the training data. The warping window (WW) intuitively controls the amount of distortion allowed when comparing a pair of time series. With a training database of N time series of lengths L, a naive approach to learning the WW requires Θ(N 2 •L 3 ) operations. This often results in NN-DTW requiring days for training on datasets containing a few thousand time series only. In this paper, we introduce FastWWSearch: an efficient and exact method to learn WW. We show on 86 datasets that our method is always faster than the state of the art, with at least one order of magnitude and up to 1000x speed-up.

Epilepsy is the most common neurological disorder. The diagnosis commonly requires manual visual electroencephalogram (EEG) analysis which is time-consuming. Deep learning has shown promising performance in detecting interictal epileptiform discharges (IED) and may improve the quality of epilepsy monitoring. However, most of the datasets in the literature are small (n≤100) and collected from single clinical centre, limiting the generalization across different devices and settings. To better automate IED detection, we cross-evaluated a Resnet architecture on 2 sets of routine EEG recordings from patients with idiopathic generalized epilepsy collected at the Alfred Health Hospital and Royal Melbourne Hospital (RMH). We split these EEG recordings into 2s windows with or without IED and evaluated different model variants in terms of how well they classified these windows. The results from our experiment showed that the architecture generalized well across different datasets with an AUC score of 0.894 (95% CI, 0.881–0.907) when trained on Alfred’s dataset and tested on RMH’s dataset, and 0.857 (95% CI, 0.847–0.867) vice versa. In addition, we compared our best model variant with Persyst and observed that the model was comparable.