OnlineSTL

Mishra, Abhinav; Sriharsha, Ram; Zhong, Sichen

doi:10.14778/3523210.3523219

Cited by 3 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we propose an accurate and efficient online STD algorithm OneShotSTL that can decompose time series online with an update complexity of 𝑂 (1). It can significantly reduce the processing time complexity of the existing methods that often require 𝑂 (𝑇 ), e.g., OnlineSTL [28]. Resultantly, the online update time of a single time point using OneShotSTL is more than 1, 000 times faster than the batch STD methods, with accuracy comparable with the best counterparts, e.g., RobustSTL [40].…”

Section: Discussionmentioning

confidence: 99%

“…Batch STD has been extensively studied in economics and finance for decades, with a variety of algorithms proposed in the literature [16,18,19,28,40]. Among them, the most popular one is STL [16], which adopts an alternating algorithm to estimate the trend and seasonal components using LOESS (LOcal regrESSion) smoothing.…”

Section: Existing Std Methodsmentioning

confidence: 99%

“…Batch or Trend Seasonality Online Online Change Shifts Complexity STL [16] Batch No No -TATBS [18] Batch No No -STR [19] Batch No No -RobustSTL [40] Batch Yes Yes -FastRobustSTL [42] Batch Yes Yes -OnlineRobustSTL [7] Online Yes Yes O(T) OnlineSTL [28] Online comparable to RobustSTL [40] (the best batch method that we are aware of in the literature) and in the meanwhile can greatly reduce the processing time. Extensive experiments on both TSAD and TSF tasks demonstrate the superior performance of OneShotSTL.…”

Section: Algorithmmentioning

confidence: 99%

“…However, most of the existing STD methods, e.g., STL [16] and RobustSTL [40], are based on batch processing, and thus essentially offline by nature, which does not support incremental updates. The only exception is a recent method OnlineSTL [28], which decomposes time series in an online fashion by effectively utilizing the fast computation from explicit mathematical closed-form expressions. Remarkably, OnlineSTL can process more than 100 times faster than traditional STD methods.…”

Section: Introductionmentioning

confidence: 99%

“…OnlineSTL[28], we divide the online processing into two phases, a single offline initialization phase and a lasting online phase. Formally, at time 𝑡 we observe {𝑦 1 , 𝑦 2 , ..., 𝑦 𝑡 0 , 𝑦 𝑡 0 +1 , ..., 𝑦 𝑡 }.…”

mentioning

confidence: 99%

See 4 more Smart Citations

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

Tan

et al. 2023

Proc. VLDB Endow.

View full text Add to dashboard Cite

Seasonal-trend decomposition is one of the most fundamental concepts in time series analysis that supports various downstream tasks, including time series anomaly detection and forecasting. However, existing decomposition methods rely on batch processing with a time complexity of O ( W ), where W is the number of data points within a time window. Therefore, they cannot always efficiently support real-time analysis that demands low processing delay. To address this challenge, we propose OneShotSTL, an efficient and accurate algorithm that can decompose time series online with an update time complexity of O (1). OneShotSTL is more than 1, 000 times faster than the batch methods, with accuracy comparable to the best counterparts. Extensive experiments on real-world benchmark datasets for downstream time series anomaly detection and forecasting tasks demonstrate that OneShotSTL is from 10 to over 1, 000 times faster than the state-of-the-art methods, while still providing comparable or even better accuracy.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Existing Std Methodsmentioning

confidence: 99%

Section: Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

Tan

et al. 2023

Proc. VLDB Endow.

View full text Add to dashboard Cite

show abstract

Gradual Leak Detection in Water Distribution Networks Based on Multistep Forecasting Strategy

Wan

Farmani

Keedwell

2023

J. Water Resour. Plann. Manage.

View full text Add to dashboard Cite

With the availability of real-time monitoring data, leakage detection for water distribution networks (WDNs) based on data-driven methods has received increasing attention in recent years. Accurate forecasts based on historical data could provide valuable information about the condition of the WDN, and abnormal events could be detected if the observed behaviour is substantially different from the typical behaviour. Therefore, an accurate forecast model is essential for prediction-based leakage detection methods. While most data-driven methods focus on burst detection, it is also important to develop an early warning system for gradual leakage events as they will cause more water loss due to a longer time to awareness. Therefore, a real-time early leakage detection technique based on a multistep forecasting strategy is proposed in this study. A multistep flow forecasting model is introduced to capture the diurnal, weekly and seasonal patterns in the historical data. The generated multistep forecasting is further compared with the observed measurements, and residuals are calculated based on cosine distance. Based on the analysis of the residual vector, the gradual leakage event could be ManuscriptClick here to access/download;Manuscript;manuscript_review_version_2 detected in a timely manner. The proposed method is applied to the L-town datasets containing one year of real-life flow monitoring data. The results prove the superiority of the proposed multistep prediction model-based method over the traditional one-step prediction model for gradual leakage detection. In addition, the results show that the proposed methodology can detect small gradual leakage events within just a few days while generating no false alarms.The method is further applied to a real-life network and showed consistent results.

show abstract

PASS: Predictive Auto-Scaling System for Large-scale Enterprise Web Applications

Guo,

Ge,

Guo

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

OnlineSTL

Cited by 3 publications

References 18 publications

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

Gradual Leak Detection in Water Distribution Networks Based on Multistep Forecasting Strategy

PASS: Predictive Auto-Scaling System for Large-scale Enterprise Web Applications

Contact Info

Product

Resources

About