A Hybrid Deep Learning Framework for Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data

dc.contributor.author Karadayı, Yıldız
dc.contributor.author Aydın, Mehmet Nafiz
dc.contributor.author Aydın, Mehmet Nafiz
dc.contributor.author Öğrenci, Arif Selçuk
dc.contributor.other Management Information Systems
dc.date.accessioned 2020-12-01T14:17:33Z
dc.date.available 2020-12-01T14:17:33Z
dc.date.issued 2020
dc.department Fakülteler, İşletme Fakültesi, Yönetim Bilişim Sistemleri Bölümü en_US
dc.department Fakülteler, Mühendislik ve Doğa Bilimleri Fakültesi, Elektrik-Elektronik Mühendisliği Bölümü en_US
dc.description.abstract Multivariate time-series data with a contextual spatial attribute have extensive use for finding anomalous patterns in a wide variety of application domains such as earth science, hurricane tracking, fraud, and disease outbreak detection. In most settings, spatial context is often expressed in terms of ZIP code or region coordinates such as latitude and longitude. However, traditional anomaly detection techniques cannot handle more than one contextual attribute in a unified way. In this paper, a new hybrid approach based on deep learning is proposed to solve the anomaly detection problem in multivariate spatio-temporal dataset. It works under the assumption that no prior knowledge about the dataset and anomalies are available. The architecture of the proposed hybrid framework is based on an autoencoder scheme, and it is more efficient in extracting features from the spatio-temporal multivariate datasets compared to the traditional spatio-temporal anomaly detection techniques. We conducted extensive experiments using buoy data of 2005 from National Data Buoy Center and Hurricane Katrina as ground truth. Experiments demonstrate that the proposed model achieves more than 10% improvement in accuracy over the methods used in the comparison where our model jointly processes the spatial and temporal dimensions of the contextual data to extract features for anomaly detection. en_US
dc.identifier.citationcount 16
dc.identifier.doi 10.3390/app10155191 en_US
dc.identifier.issn 2076-3417 en_US
dc.identifier.issn 2076-3417
dc.identifier.issue 15 en_US
dc.identifier.scopus 2-s2.0-85088858134 en_US
dc.identifier.scopusquality Q2
dc.identifier.uri https://hdl.handle.net/20.500.12469/3498
dc.identifier.uri https://doi.org/10.3390/app10155191
dc.identifier.volume 10 en_US
dc.identifier.wos WOS:000568130900001 en_US
dc.institutionauthor Aydın, Mehmet Nafiz en_US
dc.institutionauthor Öǧrenci, Arif Selçuk en_US
dc.language.iso en en_US
dc.publisher Mdpi en_US
dc.relation.journal Applıed Scıences-Basel en_US
dc.relation.publicationcategory Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı en_US
dc.rights info:eu-repo/semantics/openAccess en_US
dc.scopus.citedbyCount 31
dc.subject Spatio-temporal anomaly detection en_US
dc.subject Unsupervised learning en_US
dc.subject Multivariate data en_US
dc.subject Deep learning en_US
dc.subject CNN en_US
dc.subject LSTM en_US
dc.subject Hurricane tracking en_US
dc.subject Hurricane Katrina en_US
dc.title A Hybrid Deep Learning Framework for Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data en_US
dc.type Article en_US
dc.wos.citedbyCount 19
dspace.entity.type Publication
relation.isAuthorOfPublication a66a9279-fa0c-4915-816f-40c93cee4747
relation.isAuthorOfPublication.latestForDiscovery a66a9279-fa0c-4915-816f-40c93cee4747
relation.isOrgUnitOfPublication ff62e329-217b-4857-88f0-1dae00646b8c
relation.isOrgUnitOfPublication.latestForDiscovery ff62e329-217b-4857-88f0-1dae00646b8c

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
A Hybrid Deep Learning Framework for Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data.pdf
Size:
4.49 MB
Format:
Adobe Portable Document Format
Description: