Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data Using Deep Learning: Early Detection of COVID-19 Outbreak in Italy
Abstract
Unsupervised anomaly detection for spatio-temporal data has extensive use in a wide variety of applications such as earth science, traffic monitoring, fraud and disease outbreak detection. Most real-world time series data have a spatial dimension as an additional context which is often expressed in terms of coordinates of the region of interest (such as latitude - longitude information). However, existing techniques are limited to handle spatial and temporal contextual attributes in an integrated and meaningful way considering both spatial and temporal dependency between observations. In this paper, a hybrid deep learning framework is proposed to solve the unsupervised anomaly detection problem in multivariate spatio-temporal data. The proposed framework works with unlabeled data and no prior knowledge about anomalies are assumed. As a case study, we use the public COVID-19 data provided by the Italian Department of Civil Protection. Northern Italy regions' COVID-19 data are used to train the framework; and then any abnormal trends or upswings in COVID-19 data of central and southern Italian regions are detected. The proposed framework detects early signals of the COVID-19 outbreak in test regions based on the reconstruction error. For performance comparison, we perform a detailed evaluation of 15 algorithms on the COVID-19 Italy dataset including the state-of-the-art deep learning architectures. Experimental results show that our framework shows significant improvement on unsupervised anomaly detection performance even in data scarce and high contamination ratio scenarios (where the ratio of anomalies in the data set is more than 5%). It achieves the earliest detection of COVID-19 outbreak and shows better performance on tracking the peaks of the COVID-19 pandemic in test regions. As the timeliness of detection is quite important in the fight against any outbreak, our framework provides useful insight to suppress the resurgence of local novel coronavirus outbreaks as early as possible.
Source
Ieee AccessVolume
8Pages
164155-164177Collections
Keywords
Anomaly detectionMachine learning
Spatial databases
Clustering algorithms
Diseases
Time series analysis
Data models
Spatio-temporal anomaly detection
Deep learning
COVID-19
Outbreak detection
Italy
Multivariate data
Unsupervised
Share
Related items
Showing items related by title, author, creator and subject.
-
Anomaly detection in walking trajectory [Yürüyüş yörüngesinde anormallik algılama]
Authors:Öǧrenci, Arif Selçuk
Publisher and Date:(Institute of Electrical and Electronics Engineers Inc., 2018)Analysis of the walking trajectory and the detection of anomalies in this trajectory provide important benefits in the fields of health and security. In this work two methods to detect anomalies in trajectories are compared. Firstly an unsupervised method is used where the conformance among trajectories are taken into consideration. Trajectories that deviate from others are qualified as anomalies. Secondly the points in the trajectories are considered as a time series. Artifical neural networks ...
-
Joint phase noise estimation and source detection [Ortak faz gürültüsü kestirimi ve kaynak sezimlemesi]
Rapidly time-varying and random disturbing effects on the phase of a signal waveform are known as phase noise. In this paper we consider the problem of joint detection of continuous-valued information source output and estimation of a phase noise by using expectation maximization (EM) algorithm. In order to estimate phase noise initial phase noise values are determined by cubic interpolation that utilizes pilot symbols. Computer simulations are performed for the proposed algorithm and the average ...
-
Detection of fraudulent activities in mobile display advertising
Most of the marketing expenditures in mobile advertising are conducted through realtime bidding (RTB) marketplaces in which ad spaces of the sellers (publishers) are auctioned for the impression of the buyers’ (advertisers) mobile apps. One of the most popular cost models in RTB marketplaces is cost per install (CPi). in a CPi campaign publishers place mobile ads of the highest bidders in their mobile apps and are paid by advertisers only if the advertised app is installed by a user. CPi cost model ...