Feature-based comparison and generation of time series

Kegel, Lars; Hahmann, Martin; Lehner, Wolfgang

doi:10.1145/3221269.3221293

Cited by 26 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Decomposition-based time series augmentation has also been adopted and shown success in many time series related tasks, such as forecasting, clustering and anomaly detection. In [Kegel et al, 2018], authors discussed the recombination method to generate new time series. It first decomposes the time series x t into trend, seasonality, and residual based on STL [Cleveland et al, 1990] x t = τ t + s t + r t , t = 1, 2, ...N…”

Section: Decomposition-based Methodsmentioning

confidence: 99%

“…Authors in [Smyl and Kuber, 2016] use samples of parameters and forecast paths calculated by a statistical algorithm called LGT (Local and Global Trend). More recently, in [Kang et al, 2019] researchers use of mixture autoregressive (MAR) models to simulate sets of time series and investigate the diversity and coverage of the generated time series in a time series feature space.…”

Section: Model-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Time Series Data Augmentation for Deep Learning: A Survey

Wen,

Sun,

Yang

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep learning performs remarkably well on many time series analysis tasks recently. The superior performance of deep neural networks relies heavily on a large number of training data to avoid overfitting. However, the labeled data of many realworld time series applications may be limited such as classification in medical time series and anomaly detection in AIOps. As an effective way to enhance the size and quality of the training data, data augmentation is crucial to the successful application of deep learning models on time series data. In this paper, we systematically review different data augmentation methods for time series. We propose a taxonomy for the reviewed methods, and then provide a structured review for these methods by highlighting their strengths and limitations. We also empirically compare different data augmentation methods for different tasks including time series anomaly detection, classification and forecasting. Finally, we discuss and highlight future research directions, including data augmentation in timefrequency domain, augmentation combination, and data augmentation and weighting for imbalanced class.

show abstract

Section: Decomposition-based Methodsmentioning

confidence: 99%

Section: Model-based Methodsmentioning

confidence: 99%

Time Series Data Augmentation for Deep Learning: A Survey

Wen,

Sun,

Yang

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In [26] , the authors present a similarity measure that studies generation methods for general time series features. Their work also presents a feature-based time series generation approach that evolves cross-domain time series datasets.…”

Section: The Need For Reliable Time Series Datamentioning

confidence: 99%

A Methodology for Validating Diversity in Synthetic Time Series Generation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…For example, the studies [31,32] [34] to synthesize energy consumption time series. A Markov chain is a stochastic process with a number of states, where a state may change to another state with a certain probability depending on one or more of the past states [35]. Inspired by these works, in this paper we design two methods based on data analysis to simulate energy consumption time series.…”

Section: Energy Consumption Synthesis Approachesmentioning

confidence: 99%

Two approaches for synthesizing scalable residential energy consumption data

Liu

Iftikhar

Huo

et al. 2019

Future Generation Computer Systems

View full text Add to dashboard Cite

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

show abstract

Feature-based comparison and generation of time series

Cited by 26 publications

References 16 publications

Time Series Data Augmentation for Deep Learning: A Survey

Time Series Data Augmentation for Deep Learning: A Survey

A Methodology for Validating Diversity in Synthetic Time Series Generation

Two approaches for synthesizing scalable residential energy consumption data

Contact Info

Product

Resources

About