Detection and Exploration of Outlier Regions in Sensor Data Streams
ICDMW '08 Proceedings of the 2008 IEEE International Conference on Data Mining Workshops
DBOD-DS: distance based outlier detection for data
DEXA'10 Proceedings of the 21st international conference on Database and expert systems applications: Part I
Anomaly localization for network data streams with graph joint sparse PCA
Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining
Research issues in outlier detection for data streams
ACM SIGKDD Explorations Newsletter
Hi-index | 0.00 |
Sensor networks play a central role in applications that monitor variables in geographic areas such as the traffic volume on roads or the temperature in the environment. A key feature users are often interested in when employing such systems is the detection of unusual phenomena, that is, anomalous values measured by the sensors. In this demonstration, we present a system, called ORDEN, that allows for the detection and (visual) exploration of outliers and anomalous events in sensor networks in real-time. In particular, the system constructs outlier regions from anomalous sensor measurements to provide for a comprehensive description of the spatial extent of phenomena of interest. With our system, users can interactively explore displayed outlier regions and investigate the heterogeneity within individual regions using different parameter and threshold settings. Using real-world sensor data streams from different application domains, we demonstrate the effectiveness and utility of our system.