Highly Comparative Feature-Based Time-Series Classification

By Ben D. Fulcher, Nick S. Jones. 2014

358 citations recorded by the source corpus.

A highly comparative, feature-based approach to time series classification is introduced that uses an extensive database of algorithms to extract thousands of interpretable features from time series. These features are derived from across the scientific time-series analysis literature, and include summaries of time series in terms of their correlation structure, distribution, entropy, stationarity, scaling properties, and fits to a range of time-series models. After computing thousands of features for each time series in a training set, those that are most informative of the class structure are selected using greedy forward feature selection with a linear classifier. The resulting feature-based classifiers automatically learn the differences between classes using a reduced number of time-series properties, and circumvent the need to calculate distances between time series. Representing time series in this way results in orders of magnitude of dimensionality reduction, allowing the method to perform well on very large data sets containing long time series or time series of different lengths. For many of the data sets studied, classification performance exceeded that of conventional instance-based classifiers, including one nearest neighbor classifiers using euclidean distances and dynamic time warping and, most importantly, the features selected provide an understanding of the properties of the data set, insight that can guide further scientific investigation.

Subfields: Artificial Intelligence, Economics and Econometrics, Signal Processing

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

Forecasting Across Time Series Databases using Long Short-Term Memory Networks on Groups of Similar Series (2017, methodology): “Following the first approach, Fulcher and Jones (2014) introduce an automated feature construction process, using a large database of time series operations, which allows them to construct over 9000 different features, representing a wide range of properties of time series.”
catch22: CAnonical Time-series CHaracteristics (2019, methodology): “While some applications may be able to justify the computational expense of searching across a large feature library such as hctsa [13,14], the availability of an eﬃcient, reduced set of features, as catch22 will make the advantages of feature-based time-series analysis more widely accessible.”
The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances (2016, background): “Fulcher and Jones’s feature-based linear classifier (FBL) (Fulcher and Jones 2014).”
Large-Scale Unusual Time Series Detection (2015, background): “The features identified in our research add to an already existing set of well established features that describe time series [4] including measures of trend, seasonality, and serial correlation [20] and spectral entropy [5].”
Forecasting: theory and practice (2020, background): “Instead, they are based on the time series features measured using different statistical operations that differentiate anomalous instances from typical behaviours (Fulcher and Jones, 2014).”
Using dynamic time warping distances as features for improved time series classification (2016, methodology): “For example, while the method by Fulcher and Jones (2014) searches among thousands of various types of features, it does not include DTW as features.”