• Associate Sponsor

Connecting the dots

Social network analysis is witnessing a surge in activity at the Indian Institute of Science,Bangalore,with work involving data from Enron,Twitter and even terror networks in the mix.

Written by Johnson T A | Published: September 4, 2011 12:17 am

From Facebook to Twitter,from business organisations to terrorists,the study of networks can often tell tales about the invisible dynamics of groups—who the leaders are,the degrees of interaction,who is close to whom,the chains of command. It can even predict impending crises.

The analysis of social networks—a confluence of the computing sciences and sociology—to unlock the hidden stories behind networks predates networking of the Facebook kind by nearly four decades.

Premised on a 1967 social psychology experiment by Stanley Milgram that showed there are six degrees of separation between two individuals—known as the small world problem—social network analysis has spawned algorithms that have been utilised to look at well-known problems like the 2001 organisational meltdown at Enron and the 9/11 al-Qaeda terrorist network.

Under-utilised in India,social network analysis has over the last couple of years seen a surge in activity at the Indian Institute of Science,Bangalore,with work involving data from Enron,Twitter and even terror networks in the mix. For instance,a 2008 thesis for the MSc (engineering) programme at the Supercomputer Education and Research Centre at IISc by Kolli Naimisha,titled ‘Applications of Social Network Analysis to Community Dynamics’,focused on using publicly available Enron email data to explore “the possibility of detecting a crisis in an organisation.”

“Our work is primarily focused on trying to understand models for early warning indicators for crisis detection in organisations using social network metrics,” Naimisha said in an email. The thesis demonstrated that the study of email patterns within a hierarchical organisation can be used to detect a crisis. It also looked at research publications from IISc for a 10-year period and used details of collaborations between authors of papers to identify evolving areas of research and new collaboration communities in the institute. The work provided a methodology that it said could be applied to studying how criminal networks like terrorist groups form,evolve,identify dominant members and to help counter terrorism.

As an offshoot of the study,former police officer and head of the Research and Analysis Wing,PKH Tharakan,a member of the national security advisory board until recently,has been trying to use methodologies evolved at IISc to study the networking patterns of known terrorist groups. “Some work has been done with data of collaborations that emerged from the arrest of Raziuddin Nasir (a youth whose arrest in 2008 unravelled a wide network of operatives). A paper is likely to be published. We are keen to look at networks like the Indian Mujahideen’s as well to understand its dynamics,” Tharakan said.

Social networking analysis gained prominence as a tool for understanding terrorist networks in 2001 when Valdis Krebs,an American sociologist involved with studying networks in business organisations,adopted some of his methods to study the network involved in the 9/11 attacks in the US. Using what are known as “network centrality metrics—degrees,closeness,and between-ness”,Krebs showed that the roles of various players in the network could be identified to narrow in on the central figures in the network. Degrees,for instance,revealed the levels of separation between two operatives,closeness measured an operative’s ability to access others in the network and monitor events while between-ness revealed the control over the flow in the network that an operative enjoyed.

Apart from using social network analysis for crisis solutions,IISc is also looking at how it can be used by corporate marketing. Researchers from the Department of Computer Science and Automation at IISc have developed game theory-based algorithms for finding influential individuals in social networks for viral marketing,determining communities for targeted advertising and business intelligence and understanding the patterns behind the evolution of different social networks. “Presently we are looking at Twitter and other social network sites data to understand privacy and influential communities,” said Associate Director of IISc,N Balakrishnan,from the Supercomputer Education and Research Centre.

For all the latest News Archive News, download Indian Express App

  1. No Comments.