A new algorithm uses the location of
blasts (red) from improvised explosive
devices to deduce the hiding places of
weapons caches (yellow) in Baghdad.
kind of a case study, expert-opinion sortof field into a quantitative area,” saysAlexander Gutfraind, a mathematicianat Los Alamos National Laboratory inNew Mexico. “If you have a mathematicalmodel that can describe the structureof a terror network—and the modelworks — then you can predict the future.”Some researchers are convinced ofmath’s merits but face roadblocks inpersuading other people that calculations can aid the fight. “These days, theimpression I get is that people who oughtto support this kind of research don’t fullybelieve that mathematics can be useful,”says mathematician Jonathan Farley ofJohannes Kepler University Linz in Austria. “And their belief is so extreme thatthey’re not even willing to check it out.”But some of the new methods arebeginning to attract attention. Farley, Gutfraind and others belong to theConsortium for Mathematical Methodsin Counterterrorism, which promotesmath’s role in tackling counterterrorism and global security problems. Consortium members share methods andpapers, and meet annually to talk aboutemerging problems in their field, such ashow to quantify threats of violence, howto disrupt terror cells and how terroristsarrange themselves into groups.
Such simple maps can miss crucial features of terrorist networks. “The personwith the most links is not necessarily themost important person,” Farley says.
Carley’s research group and others have begun to find that people whohave roles in multiple groups — calledinterstitial members—are some ofthe most important. Interstitial members communicate between groups andrelay information, a position criticalto operations going as planned. A newtechnique called fuzzy grouping, whichallows people to be assigned to multiplegroups simultaneously, better describeshow these people fit into networks,Carley says.
Another important attribute is exclusivity. Some members of a network havespecialized training and so are in highdemand for certain jobs. People whoknow how to launder money or fly an airplane, for instance, have a high exclusivity measure. Accounting for exclusivitymeasures and adopting fuzzy groupingtechniques can lead to more nuanceddescriptions of covert networks.
Putting terrorists into groups or
assigning exclusivity measures is a mat-
ter of collecting and assessing the right
pieces of data. “The data is coming from a
wide variety of sources, things like open-
source text, crowd-sourced information
off the Web, anything you can basically
imagine,” says Carley. “The million dol-
lar question is, can we drill down and find
the network relevant to the problem?”
Ideally, drilling down through this
aggregation of data will reveal trails.
“We link together the who, the what,
the where, the why, the how, and we
use all of these things in a dynamic
complex configuration,” Carley says.
Once specific trails are identified, com-
plex grouping algorithms may be able
to decipher unexpected locations for
groups to meet, for instance.
Linked up a type of network arrangement called fuzzy group clustering pinpoints people whobelong to two distinct groups simultaneously, such as Saddam hussein. models suggest thatsuch “interstitial” members are likely to be the key coordinators of a terrorist mission.
Networks linking Iraqi leaders,other individuals and mosques
Muhammed SadrMuhammed Bakr
Connecting the right dotsIn the aftermath of the September 11attacks, a technique called social network analysis was touted as the best wayto find terrorist kingpins. Connecting thedots between people called attention tothose who were most highly linked — presumably, the most important membersof the network. “The idea was to disconnect those social networks, and if you didthis, you could inhibit, prevent or moderate the impact of these events — andmaybe actually save lives,” Carley says.“What we found in the ensuing time isthat taking an approach that focuses onlyon social networks will not work.”
Osama bin Laden
July 17, 2010 | science news | 19