Tracking zebras... and consumers
Wireless networks have been used before to track animals, such as elephants, koalas, and even pigeons. Now, U.S. researchers are using social networking software to track zebras. They hope to discover how animals interact -- especially when lions are near a herd. They also think their software can help ecologists devise new techniques to protect endangered species. They even think that their computational tools can easily be adapted to study our buying habits. Does this mean that we'll be fitted with a GPS tracking collar before entering a shopping mall?
You can see above a picture of a zebra with a sensor collar and a snapshot of zebra population in Kenya (Credit: Computational Tools for Population Biology). This project has been led by University of Illinois at Chicago assistant professor of computer science Tanya Berger-Wolf, director of the Laboratory for Computational Population Biology. She worked with Princeton University ecologist Daniel Rubenstein and University of New Mexico computer scientist Jared Saia.
So how these researchers will track a zebra population? "The initial focus is on zebras living in Kenya's vast Mpala conservancy. A number of the animals will be fitted with GPS tracking collars that will provide researchers with a more accurate picture of life among the herd, showing how animals interact and which one leads the herd to flee when predators, notably lions, are near. Zebra species have mainly been reduced to three -- the endangered Grevy's, the Mountain and the common Plains. Rubenstein has studied zebras for more than two decades and hopes to learn more from tracking the social habits of the Plains and Grevy's species to see how they differ in evading predators."
And how this will really work? "The zebras will be tracked every 8 to 15 minutes, and the data will be relayed by cell phone to the researchers' computers, where new computational and analytical software tools developed as part of the project will help map and analyze the animals' social networking in ways never done before. The tools will help researchers study the time and order of animal social interaction. The approach combines ideas from academic disciplines such as social network analysis, Internet computing, data mining and machine learning to solve the complicated puzzles of population biology."
Speaking about computational population biology, here is how Tanya Berger-Wolf describes her research. "Flu pandemic, political microtargeting, behavioral response to predator presence, species genetic diversity. Populations contain intricate connections that occur on time scales ranging from milliseconds to generations. At the Laboratory for Computational Population Biology, we explore the growing interface between Population Biology and Computer Science, from genetics to social interactions."
And here is her description of Dynamic Network Analysis. "Finding patterns of social interaction within a population has applications from epidemiology and marketing to conservation biology and behavioral ecology. An intrinsic characteristics of societies is their continual change. Yet, few analysis methods are explicitly dynamic. We are working on novel conceptual and computational frameworks to accurately describe the social context of an individual at time scales matching changes in individual and group activity. Current projects in this direction include finding communities and critical individuals in dynamic networks, and fine-grained interaction prediction in dynamic networks."
For more information, here are three documents I recommend.
- Reconstructing Sibling Relationships in Wild Populations (PDF format, 9 pages), which will soon appear in Bioinformatics
- A computational approach to animal breeding (PDF format, 10 pages), which has been published in the Journal of Theoretical Biology (Volume 244, Issue 3, Pages 433-439, February 7, 2007, link to abstract)
- Multichannel Communication and Graph Vertex Labeling (Powerpoint format, 16 slides), a presentation which includes very interesting images about social networks
Still, I don't know how the researchers want to study our buying habits without equipping us with special devices. After all, we don't all carry GPS collars...
Sources: University of Illinois at Chicago news release, September 6, 2007; and various websites
You'll find related stories by following the links below.