Finding Spatio-Temporal Patterns in Multidimensional Data Streams


  • Santiago A. Nunes University of São Paulo
  • Luciana A. S. Romani Embrapa Agricultural Informatics - Campinas
  • Ana M. H. Avila Cepagri - University of Campinas
  • Priscila P. Coltri Cepagri - University of Campinas
  • Agma J. M. Traina University of São Paulo
  • Elaine P. M. Sousa University of São Paulo



spatio-temporal data, fractals, multi-resolution spatial structure


In the last few decades, advances in data acquisition technology have contributed to generation of huge volumes of data in diverse application areas, creating new research challenges in knowledge discovery.The analysis of these data has become an important task in several domains such as sensor networks, web-logs, financial transactions and climate change monitoring. In this article, we propose the Spatio-Temporal Behavior Meter (STB-meter) method to identify spatio-temporal patterns in multidimensional evolving data streams. Our approach combines a multi-resolution hierarchical structure to deal with spatial information with fractal-based analysis to monitor non spatial information of the multidimensional data stream. Experimental evaluation on real climate data shows that our method allows finding relevant spatio-temporal patterns in evolving data at different spatial and temporal resolutions and therefore it can be a useful tool to assist domain specialists in climate change researches.


Download data is not yet available.




How to Cite

Nunes, S. A., Romani, L. A. S., Avila, A. M. H., Coltri, P. P., Traina, A. J. M., & Sousa, E. P. M. (2013). Finding Spatio-Temporal Patterns in Multidimensional Data Streams. Journal of Information and Data Management, 4(3), 327.



SBBD Articles