Abstract

Thanks to the advance of novel technologies, such as sensors and Internet of Things (IoT) technologies, big amounts of data are continuously gathered over time, resulting in a variety of time series. A semi-supervised anomaly detection framework, called Tri-CAD, for univariate time series is proposed in this paper. Based on the Pearson product-moment correlation coefficient and Dickey–Fuller test, time series are first categorized into three classes: (i) periodic, (ii) stationary, and (iii) non-periodic and non-stationary time series. Afterwards, different mechanisms using statistics, wavelet transform, and deep learning autoencoder concepts are applied to different classes of time series for detecting anomalies. The performance of the proposed Tri-CAD framework is evaluated by experiments using three Numenta anomaly benchmark (NAB) datasets. The performance of Tri-CAD is compared with those of related methods, such as STL, SARIMA, LSTM, LSTM with STL, and ADSaS. The comparison results show that Tri-CAD outperforms the others in terms of the precision, recall, and F1-score.

Highlights

  • Detection involves identifying data that do not conform to the pattern as expected [1,2]

  • This paper focuses on univariate time series and proposes a semi-supervised anomaly detection framework, called Tri-CAD, for such time series

  • This is because related methods try to improve anomaly detection performance by integrating various models, each of which is suitable for a different class of time series

Read more

Summary

Introduction

Detection involves identifying data that do not conform to the pattern as expected [1,2]. Tri-CAD is semi-supervised in the sense that it takes the first D (say, D = 1000) data of time series as input under the assumption that the first D data are normal. The contribution of this research is threefold It proposes the Tri-CAD framework to perform anomaly detection by first classifying time series into different classes, and applying different anomaly detection methods that are suitable for different classes of data. It implements the Tri-CAD framework and applies the implementation to real-world datasets to validate the concept of the framework.

Time Series and Related Models
Pearson Correlation Coefficient
Dickey–Fuller Test
Autoencoder
Proposed
Periodic Time Series Anomaly Detection
StationarytoTime
Non-Periodic and Non-Stationary Time Series Anomaly Detection
Treshold
Predecesors of Tri-CAD
Performance Evaluation and Comparison
Tri-CAD
10. Tri-CAD
11. Tri-CAD
13. Tri-CAD performance evaluation using using
14. Tri-CAD
Findings
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call