Emerging scientific applications in data mining
Communications of the ACM - Evolving data mining into solutions for insights
On the Generation of Spatiotemporal Datasets
SSD '99 Proceedings of the 6th International Symposium on Advances in Spatial Databases
On detecting space-time clusters
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
The hunting of the bump: on maximizing statistical discrepancy
SODA '06 Proceedings of the seventeenth annual ACM-SIAM symposium on Discrete algorithm
Spatial scan statistics: approximations and performance study
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
ICDMW '07 Proceedings of the Seventh IEEE International Conference on Data Mining Workshops
Mote-Based Online Anomaly Detection Using Echo State Networks
DCOSS '09 Proceedings of the 5th IEEE International Conference on Distributed Computing in Sensor Systems
Discovering spatio-temporal causal interactions in traffic data streams
Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining
Mining trajectories of moving dynamic spatio-temporal regions in sensor datasets
Data Mining and Knowledge Discovery
Hi-index | 0.00 |
The detection of outliers from spatio-temporal data is an important task due to the increasing amount of spatio-temporal data available and the need to understand and interpret it. Due to the limitations of current data mining techniques, new techniques to handle this data need to be developed. We propose a spatio-temporal outlier detection algorithm called Outstretch, which discovers the outlier movement patterns of the top-k spatial outliers over several time periods. The top-k spatial outliers are found using the Exact-Grid Top-k and Approx-Grid Top-k algorithms, which are an extension of algorithms developed by Agarwal et al. [1]. Since they use the Kulldorff spatial scan statistic, they are capable of discovering all outliers, unaffected by neighbouring regions that may contain missing values. After generating the outlier sequences, we show one way they can be interpreted, by comparing them to the phases of the El Niño Southern Oscilliation (ENSO) weather phenomenon to provide a meaningful analysis of the results.