Penn State, Distinguished Professor features workshop at dg.o 2006
By Karen Heyman
For the DGRC
At this year’s 7th Annual International Conference on Digital Government Research, there will be an all-day workshop on “Digital Governance and Hotspot GeoInformatics for Monitoring, Etiology, Early Warning, and Sustainable Management.” Taking place on Sunday, May 21, 2006 in San Diego, it will be conducted by Ganapati P. Patil, Distinguished Professor and Director, Penn State Center for Statistical Ecology and Environmental Statistics.
Geoinformatics is a relatively new discipline that uses the tools of computer science to analyze large sets of geographically based data. It is especially relevant to practitioners in public health and public policy for issues ranging from persistent poverty to epidemics and the aftermath of natural disasters. It is also valuable for those in varied environmental management disciplines, from conservation to invasive species management. Additionally, it can provide critical information for security professionals for object recognition and tracking, and for potential spread of the effects of bioterror weapons.
The workshop will cover geoinformatic models and tools appropriate for detecting and prioritizing “hotspots.” Patil and his colleagues have been working on how to identify potential "hotspots." These can be crisis points in the military sense involving areas and networks of robots, sensors, or wireless devices, but also in the environmental and human health sense involving events of societal importance over geographic regions or across networks.
To take a public health example, a “hotspot” could refer to where a disease outbreak has started, where it is at its worst, and most importantly, where and when one may emerge. A public policy analyst might try to detect a hotspot of poverty, looking at both geography and temporal dynamics, as Patil explains, “The questions to ask are, ‘What are the poverty patches in the city, how have they changed over time, have the patches been growing, spreading, merging, shifting? Are there any poverty patch trajectories that can be useful for making policy for poverty alleviation’?”
The identification process for these hotspots requires the coordination of many spatial and temporal parameters. In order to correctly incorporate and interpret all of these parameters, Patil and his collaborators start with the spatial scan statistic, a widely used metric in public health, and modify it to apply to environmental sciences and elsewhere.
"The popular health-area scan statistic is a statistical method designed to detect a local excess of events and to test if such an excess can reasonably have occurred by chance," explains Patil. "However, its major limitation is that it is circle-based. The clusters can be of any shape, and cannot be captured only by circles. In more general settings, this is likely to give more false alarms and more of a false sense of security. What we need is the capability to detect arbitrarily shaped hotspots. We plan to accomplish this using our innovation with upper level sets and their connected components extended to permutation based upper level sets (PULSE), and also using novel genetic algorithms."
The benefit of hotspot detection is that it allows for customized analysis and thus the ability to create more specific, targeted policies. “One can see that while 400 cities don’t need 400 policies, they cannot do with one single policy, one size fits all,” says Patil, “Unfortunately, that has been the case so far in most countries when it comes to poverty alleviation, and by and large the policies have failed.”
Detection is only the first step, equally important is prioritizing. “Once you identify hotspots, you want to know which hotspots need monitoring, which ones can be used for etiological analysis, which ones are ready for some assessment and management, so you want to prioritize and rank these hotspots.” Of course, as practitioners know, there are different stakeholders in every situation. With each stakeholder providing a score for each hotspot, “You don’t want to construct an index, because it’s a question of apples and oranges,” says Patil, “You want to prioritize and rank these hotspots without crunching these multiple indicators into a single index, so the question is how this can be done.”
A conventional solution is to assign a composite numerical score to each hotspot by combining the indicator information in some fashion. But warns Patil, “Consciously or otherwise, every such composite involves judgments, often arbitrary or controversial about tradeoffs or substitutability among indicators.”
In Patil’s methodology, the solution is, “Rather than trying to combine indicators, we take the view that the relative positions in indicator space determine only a partial ordering and that a given pair of hotspots may not be inherently comparable. Working with Hasse diagrams of the partial order, we study the collection of all rankings that are compatible with the partial order. And using our innovation with partial order sets, rank frequency distributions, and cumulative rank frequency operators, we accomplish the desired prioritization of the hotspots.”
In essence, it is a problem in multi-dimensional space. You can do an x, y plot to measure say, income by age, but suppose you have two hundred countries and you want to rank them according to human/environmental interface progress, which includes several indicators. The Hasse diagram allows you to collapse several dimensions into levels that can be compared against each other, and provides a basis for insightful analysis leading to a broadly acceptable ranking.
Says Patil, “We will share these and other methods and tools with the workshop in the hopes that the participants will be able to use some of these methods and tools for analyzing their own datasets and databases if they’re looking for hotspots and prioritizing them.”
WORKSHOP STRUCTURE
The workshop will be divided into morning and afternoon sessions, with the morning devoted to the toolbox exposition with examples and datasets, with particular emphasis on public health, crime, ecohealth, poverty, networks and the like. In the afternoon, there will be actual case studies presented by people working around the world. “One panel participant is an Indian New Delhi official in the Department of Information Technology, who is a senior director of the e-government program,” says Patil. “An innovative young expert from Brazil is going to present a hotspot case study on dengue fever and malaria in Brazil, using creative genetic algorithms together with environmentally meaningful adjacency definitions. My colleague from Penn State will make a case study on crop disease and the management advisory on whether a farmer should spray or not spray. A vice-president of the famous Bogor Agricultural University in Indonesia is going to speak about what Indonesia is going to do to implement hotspot geoinformatics.”
Patil welcomes participants to submit their own case studies for presentation: “It will be a good workshop for people to attend, and speak, and share some of their issues and datasets.” Participants are also welcome just to watch, hear, and overhear the exciting developments and applications for this century.
Links:
http://www.stat.psu.edu/hotspots
http://en.wikipedia.org/wiki/Hasse_diagram
