Over the weekend, Team Crowdscanner won the State Department’s global "Tag Challenge" contest to find pretend “suspects” in five cities: Washington DC, New York City, Stockholm, London and Bratislava. The thrust of the contest was for individuals or teams to devise innovative ways of getting people to use social network technologies to voluntarily locate and report a suspect in a city’s everyday spaces: on the street, on public transportation, in a cafe, etc. Since this is an impossible endeavor for a single individual, most teams used compensation structures to entice people into participation. Using what’s called a “recursive incentive mechanism," Team Crowdscanner devised two ways for people to contribute: (1) a money prize for an uploaded image of a suspect, and (2) a $1 recruiting prize for each recruit brought in to participate, up to $2000. Each team had a day, March 31, to find all five suspects. The MIT-based Team Crowdscanner, relying on social networks, won the contest by only finding three.
The participants involved in the Tag Challenge won a similar contest set up by DARPA a couple of years ago, where ten red balloons were placed throughout the United States, and a team at MIT found the balloons in under nine hours using web-based social networks and computational network analysis.
As I noted in another post, I’m writing an article on how this kind of research is integral to contemporary U.S. military operations. And, it should come as no surprise that the computer scientists who are winning these prizes are consultants either for the U.S. military, or are members of scientific advisory boards which contract through the military.
Given this, I thought it might be useful to write a little “Who’s who” featuring three people in the social sciences that work closely with the military to develop modelling and simulation techniques for use in so-called “battlespaces” (a category that is a clue for how late-modern warfare is conceived). This is merely a who’s who; I unfortunately do not have the time to delve precisely into how their research is utilized by the military, though I will elaborate on that in the coming weeks… What’s important to keep in mind is its this kind of research that is leading to fruit-loop ideas like developing "social radar" to predict revolutions, a serious and potentially dangerous notion in the head of the Air Force’s chief scientist Dr. Mark Maybury.
Perhaps the most important social scientist who consults the military is one of the top computer scientists in the United States, Alex “Sandy” Pentland. Named the #6 “most powerful data scientist” in the world by Forbes Magazine, Pentland directs the MIT Media Lab and the MIT Human Dynamics Laboratory, which is financially underwritten by Google, Microsoft, Nokia, the NSF, the US Army Research Lab, DARPA and others. Pentland specializes in two fields: what he calls “human signals" and "reality mining.” In an interview with Forbes Magazine, Pentland describes reality mining, which is “machine-sensed environmental data” in the following terms:
"Data mining is about finding patterns in digital stuff. I’m more interested specifically in finding patterns in humans… I’m taking data mining out into the real world… We’ve studied human behavior, and now we’re learning how to shape it."
Pentland has not only created start-ups incorporating these ideas, but he has tried to work with the U.S. military to fashion what he calls “Computational Counterinsurgency,” in order to use data mining to predict human behavior in counterinsurgency campaigns.
Pentland also serves on the advisory board for Aptima, which describes itself as a “Human-Centered Engineering Firm.” A leading scientist at the firm is the highly influential Jean MacMillan. MacMillan has worked with Pentland on research for the Navy to fabricate what they call “similarity measures for human networks.” MacMillan is renowned in military social science circles, and she was a lead author on the US Air Force commissioned book Behavioral Modeling and Simulation. In many ways, this book is an updated version of the influential 1963 manuscript, Social Science Research and National Security, edited by political scientist (and early developer of social network analysis) Ithiel de Sola Pool.
Serving on the 13-member committee who wrote Behavioral Modeling and Simulation is Kathleen Carley at the School of Computer Science and Institute for Software Research at Carnegie Mellon University. Carley’s work has been integral for developing counterterrorism computer programs based on “dynamic network analysis” for identifying terrorist cell structures. She also works on simulation systems to predict insurgent and terrorist activities.
Other social scientists could be mentioned, especially if one were loose with the term (e.g., David Kilcullen). But Pentland, MacMillan and Carley presently constitute the creme de la creme for research being developed to predict and “neutralize” insurgent activity.
Here are videos of Pentland and Carley talking about their research:
Pentland (here): Note the talk prefaces with crisis, and then utilizes “natural” ways of avoiding crisis.
Carley (here): on “assessing terror networks”
- winterruhe reblogged this from darpadreaming and added:
- darpadreaming posted this