Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests (tsfresh - A Python package)

By Maximilian Christ, Nils Braun, Julius Neuffer, Andreas W. Kempa-Liehr. Neurocomputing, 2018

1,217 citations recorded by the source corpus.

Abstract Time series feature engineering is a time-consuming process because scientists and engineers have to consider the multifarious algorithms of signal processing and time series analysis for identifying and extracting meaningful features from time series. The Python package tsfresh (Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests) accelerates this process by combining 63 time series characterization methods, which by default compute a total of 794 time series features, with feature selection on basis automatically configured hypothesis tests. By identifying statistically significant time series characteristics in an early stage of the data science process, tsfresh closes feedback loops with domain experts and fosters the development of domain specific features early on. The package implements standard APIs of time series and machine learning libraries (e.g. pandas and scikit-learn ) and is designed for both exploratory analyses as well as straightforward integration into operational data science applications.

Subfields: Artificial Intelligence, Economics and Econometrics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/82f472fb8e5f7db675f7b75000047b637b2facc7

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

1. Fotios Petropoulos
2. Christoph Bergmeir
3. Spyros Makridakis
4. Nikolaos Kourentzes
5. Vassilios Assimakopoulos
6. Devon K. Barrow
7. Yanfei Kang
8. Feng Li

In Economics and Econometrics

1. Fotios Petropoulos
2. Konstantinos Nikolopoulos
3. Rob J Hyndman
4. Nikolaos Kourentzes
5. Vassilios Assimakopoulos
6. Evangelos Spiliotis
7. Spyros Makridakis
8. Christoph Bergmeir

In Signal Processing

1. Vassilios Assimakopoulos
2. Konstantinos Nikolopoulos
3. Evangelos Spiliotis
4. Spyros Makridakis
5. Fotios Petropoulos
6. Nikolaos Kourentzes
7. Christoph Bergmeir
8. Devon K. Barrow

Citations in the corpus

• Forecasting: theory and practice (2020, methodology): “on methods quantify a wide range of time series characteristics, including simple summary statistics, stationarity (MonteroManso et al., 2020; Wang et al., 2020c), model fits (Fulcher and Jones, 2014; Christ et al., 2018), time series imaging (Li et al., 2020f), and others. In the forecasting communi”
• Tslearn, A Machine Learning Toolkit for Time Series Data (2020, background): “Similarly, tsfresh (Christ et al., 2018) specializes in feature extraction from time series. pyts (Faouzi and Janati, 2020) and sktime (L¨oning et al., 2019), on the other hand, focus on supervised learning.”
• TSFEL: Time Series Feature Extraction Library (2020, methodology): “In literature, there exist related packages dedicated to feature extraction, such as FATS [2], CESIUM [3], TSFRESH [4] and HCTSA [5].”