tegdet: An extensible Python library for anomaly detection using time evolving graphs
This paper presents tegdet, a new Python library for anomaly detection in unsupervised approaches. The input of the library is a univariate time series, representing observations of a given phenomenon. Then, tegdet identifies anomalous epochs, i.e., time intervals where the observations differ in a...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-05-01
|
| Series: | SoftwareX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352711023000596 |
| _version_ | 1827940541431545856 |
|---|---|
| author | Simona Bernardi Raúl Javierre José Merseguer |
| author_facet | Simona Bernardi Raúl Javierre José Merseguer |
| author_sort | Simona Bernardi |
| collection | DOAJ |
| description | This paper presents tegdet, a new Python library for anomaly detection in unsupervised approaches. The input of the library is a univariate time series, representing observations of a given phenomenon. Then, tegdet identifies anomalous epochs, i.e., time intervals where the observations differ in a given percentile of a baseline distribution. Epochs are represented by time evolving graphs and the baseline distribution is given by the dissimilarities between a reference graph and the graphs of the epochs. Currently, the library implements 28 dissimilarity metrics, i.e., 28 different anomaly detection techniques, and its extensible design allows to easily introduce new ones. tegdet exposes a complete functionality to carry out the anomaly detection, through a straightforward designed API. Summarizing, to the best of our knowledge, tegdet is the first publicly available library, based on time evolving graphs, for anomaly detection in time series. Our experimentation shows promising results. For example, Clark and Divergence techniques can achieve an accuracy of 100%, while the time to build the model and predict lasts for few hundreds milliseconds. |
| first_indexed | 2024-03-13T09:10:52Z |
| format | Article |
| id | doaj.art-49c452273417425b8605429bbcb47087 |
| institution | Directory Open Access Journal |
| issn | 2352-7110 |
| language | English |
| last_indexed | 2024-03-13T09:10:52Z |
| publishDate | 2023-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | SoftwareX |
| spelling | doaj.art-49c452273417425b8605429bbcb470872023-05-27T04:25:51ZengElsevierSoftwareX2352-71102023-05-0122101363tegdet: An extensible Python library for anomaly detection using time evolving graphsSimona Bernardi0Raúl Javierre1José Merseguer2Dept. de Informática e Ing. de Sistemas, Universidad de Zaragoza, Spain; Corresponding author.Hiberus, SpainDept. de Informática e Ing. de Sistemas, Universidad de Zaragoza, SpainThis paper presents tegdet, a new Python library for anomaly detection in unsupervised approaches. The input of the library is a univariate time series, representing observations of a given phenomenon. Then, tegdet identifies anomalous epochs, i.e., time intervals where the observations differ in a given percentile of a baseline distribution. Epochs are represented by time evolving graphs and the baseline distribution is given by the dissimilarities between a reference graph and the graphs of the epochs. Currently, the library implements 28 dissimilarity metrics, i.e., 28 different anomaly detection techniques, and its extensible design allows to easily introduce new ones. tegdet exposes a complete functionality to carry out the anomaly detection, through a straightforward designed API. Summarizing, to the best of our knowledge, tegdet is the first publicly available library, based on time evolving graphs, for anomaly detection in time series. Our experimentation shows promising results. For example, Clark and Divergence techniques can achieve an accuracy of 100%, while the time to build the model and predict lasts for few hundreds milliseconds.http://www.sciencedirect.com/science/article/pii/S2352711023000596Unsupervised anomaly detectionUnivariate time-seriesTime evolving graphsDissimilarity metrics |
| spellingShingle | Simona Bernardi Raúl Javierre José Merseguer tegdet: An extensible Python library for anomaly detection using time evolving graphs SoftwareX Unsupervised anomaly detection Univariate time-series Time evolving graphs Dissimilarity metrics |
| title | tegdet: An extensible Python library for anomaly detection using time evolving graphs |
| title_full | tegdet: An extensible Python library for anomaly detection using time evolving graphs |
| title_fullStr | tegdet: An extensible Python library for anomaly detection using time evolving graphs |
| title_full_unstemmed | tegdet: An extensible Python library for anomaly detection using time evolving graphs |
| title_short | tegdet: An extensible Python library for anomaly detection using time evolving graphs |
| title_sort | tegdet an extensible python library for anomaly detection using time evolving graphs |
| topic | Unsupervised anomaly detection Univariate time-series Time evolving graphs Dissimilarity metrics |
| url | http://www.sciencedirect.com/science/article/pii/S2352711023000596 |
| work_keys_str_mv | AT simonabernardi tegdetanextensiblepythonlibraryforanomalydetectionusingtimeevolvinggraphs AT rauljavierre tegdetanextensiblepythonlibraryforanomalydetectionusingtimeevolvinggraphs AT josemerseguer tegdetanextensiblepythonlibraryforanomalydetectionusingtimeevolvinggraphs |