Time series shapelets: a novel technique that allows accurate, interpretable and fast classification

By Lexiang Ye, Eamonn J. Keogh. Data mining and knowledge discovery, 2010

425 citations recorded by the source corpus.

Classification of time series has been attracting great interest over the past decade. While dozens of techniques have been introduced, recent empirical evidence has strongly suggested that the simple nearest neighbor algorithm is very difficult to beat for most time series problems, especially for large-scale datasets. While this may be considered good news, given the simplicity of implementing the nearest neighbor algorithm, there are some negative consequences of this. First, the nearest neighbor algorithm requires storing and searching the entire dataset, resulting in a high time and space complexity that limits its applicability, especially on resource-limited sensors. Second, beyond mere classification accuracy, we often wish to gain some insight into the data and to make the classification result more explainable, which global characteristics of the nearest neighbor cannot provide. In this work we introduce a new time series primitive, time series shapelets, which addresses these limitations. Informally, shapelets are time series subsequences which are in some sense maximally representative of a class. We can use the distance to the shapelet, rather than the distance to the nearest neighbor to classify objects. As we shall show with extensive empirical evaluations in diverse domains, classification algorithms based on the time series shapelet primitives can be interpretable, more accurate, and significantly faster than state-of-the-art classifiers.

Subfields: Artificial Intelligence, Economics and Econometrics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/450de7320b94e3df15eb9634ae17b1825165c4d6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Economics and Econometrics

In Signal Processing

Citations in the corpus

Mapping the stereotyped behaviour of freely moving fruit flies (2013, methodology): “For the results presented here, we look at 25 frequency channels, dyadically spaced between 1 Hz and 50 Hz, the larger of which being the Nyquist frequency of the system.”
The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances (2016, methodology): “The k-best SAX words are selected then mapped back to the original shapelets, which are assessed using information gain in a way identical to that used in Ye and Keogh (2011).”
Classification of time series by shapelet transformation (2013, methodology): “We conduct our experiments on 29 datasets: 17 from the UCR repository, and 12 we provide ourselves.”
Deep learning for time series classification: a review (2018, methodology): “Ye and Keogh (2011) identified that the most important shapelet for the Gun/NoGun classification occurs when the actor’s arm is lowered (about 120 on the horizontal axis in Figure 13).”
Time-Series Classification with COTE: The Collective of Transformation-Based Ensembles (2015, background): “The ﬁrst involves ﬁnding shapelets in the dataset [16].”
A Bag-of-Features Framework to Classify Time Series (2013, background): “As an alternative, [10], [11], [12] proposed to identify subsequences (called shapelets) that are representative of a class using euclidean distance.”
Learning time-series shapelets (2014, background): “4 Distances Between Shapelets and Series The distance between the i-th series Ti and the k-th shapelet Sk is defined as the minimum distance Mi,k (shown in Equation 1) among the distances between the shapelet Sk and each segment j of Ti [17, 18].”