Special Issue "Spatial Analysis and Data Mining"

Guest Editor
Prof. Dr. Brian Lees
School of Physical, Environmental and Mathematical Sciences, UNSW Canberra, PO Box 7916, Canberra, BC 2610, Australia
Website: http://pems.unsw.adfa.edu.au/staff/profiles/lees/index.html
E-Mail: b.lees@adfa.edu.au
Phone: +61 2 6268 8302
Fax: +61 2 6268 8017
Interests: global change; predictive mapping of land cover and land degradation

Dear Colleagues,

Traditional spatial analyses grew up in an era of sparse data and very weak computational power. Today, both of those circumstances are reversed and many of the old solutions are no longer suitable. The title of this Special Issue, "Spatial Analysis and Data Mining", reflects this change and combines two things which, until recently, engaged quite different groups of researchers and practitioners. Together, they require particular techniques and a sophisticated understanding of the special problems associated with spatial data. This geographic data mining, or Geographic Knowledge Discovery (GKD), is not new, but is developing and changing rapidly as both more, and different, data becomes available, and people see new applications. The days of ‘Big Data’ require fresh thinking.

The aim of geographic data mining (GKD) is to assist in the generation of hypotheses, which can be tested, about interesting or anomalous spatial patterns which may be discovered in very large databases. It is important that the patterns discovered should not be statistical or sampling artifacts, and should be nontrivial and useful. The intent is not to build a system that makes decisions or interpretations automatically, but supports humans in these tasks. Also GKD is not synonymous with statistical analyses, such tools have a role in the testing of hypotheses generated by GKD but not in GKD itself.

We seek original and innovative papers which address this fusion of “Spatial Analysis and Data Mining” and present research which advances theory, demonstrates application and evaluates the approach taken.

Professor Brian Lees
Guest Editor

• geographic data mining
• geographic knowledge discovery
• spatio-temporal data mining
• spatial analysis
• knowledge discovery

Type of Paper: Article
Title: A New Algorithm for Identifying Possible Epidemic Sources with Application to the German Escherichia Coli Outbreak
Authors: Massimo Buscema ^1,2,* Enzo Grossi ^1,3, Al Bronstein⁴, Weldon Lodwick²,  Masoud Asadi-Zeydabadi^2,5, Roberto Benzi⁶, Francis Newman²
Affiliations: ¹ Semeion, Research Centre of Sciences of Communication, Via Sersale n. 117, 00128 Rome, Italy; ² Dept. of Mathematical and Statistical Sciences, CCMB, University of Colorado Denver, CO, USA.; ³ Bracco Foundation, Milano, Italy; ⁴ Rocky Mountains Poison and Drug Center, Denver, CO, USA; ⁵ Dept. of Physics, University of Colorado Denver, CO, USA; ⁶ Dept. of Physics, Tor Vergata University, Rome, Italy
Abstract: In this paper we describe in detail equations generated by a recently developed algorithm called Topological Weighted Centroid (TWC).  TWC takes locations of an event of interest, for example disease outbreaks (called “assigned points” here), and analyzes the possible associated dynamics. In this algorithm the ideas of free energy and entropy have been used.   This novel mathematical tool has been applied to a real world example, the epidemic outbreak of hemolytic uremic syndrome (HUS) caused by Shiga toxin-producing Escherichia coli (E-Coli) that occurred in Germany in 2011. Applying TWC made it possible to point out the source of the outbreak.  Based on scanty data collected on May 30, 2011 and analyzed and reported on June 3, 2011, our analysis showed two sources, one near Hamburg that was already suspected and one 5 km from Frankfurt that was discovered June 18, 2011.
This novel algorithm may help decision makers in situations with limited amounts of spatial information, and reflects the power of mathematics of complex systems in improving the level of accuracy obtained with classical statistics. In particular TWC can be used as a powerful method to identify the source of epidemic spread. The impressive results obtained in the example of the 2011 German HUS epidemic are consistent with the idea that the spread of infectious disease is not random but follows a progression based on inherent, but as yet undiscovered, mathematical laws. This method, which requires further field evaluation and validation, could provide an additional powerful tool for the investigation of the early stages of an epidemic, and constitute the basis of novel simulation methods to understand the process through which a disease is spread.
Keywords: topological weighted centroid; epidemic out break; E-coli; HUS epidemics