If your Facebook page - or "node" - disappeared, would your whole social network come crashing to a halt?  
 
A five-year, $5.4 million award to UC Irvine researchers will help them find out. The grant from the U.S. Office of Naval Research will allow a research team to analyze networks of data on a significantly larger scale than ever before. The goal is to help scientists understand how networks the size of Facebook and LinkedIn are formed and how they evolve over time. The millions of Web pages connected through hyperlinks also will be studied.  
 
Padhraic Smyth, computer science professor and director of UCI's Center for Machine Learning and Intelligent Systems, is leading the study along with co-investigators Michael Goodrich and David Eppstein from UCI's Donald Bren School of Information and Computer Sciences. They will collaborate with sociology professor Carter Butts and researchers at the University of Washington, Pennsylvania State University and University of Maryland. The research team also includes a total of 12 graduate students and postdoctoral fellows from all four universities.  
Related research traditionally has focused on networks of just a few hundred connected individuals, or nodes. In this new project, UCI researchers will develop mathematical models for studying networks with millions of nodes.  
 
Modeling techniques developed through this research could be used to identify critical network nodes and predict likely areas of new network growth. Project research also could spur the development of software that automatically creates new links among people in large organizations or on the Web to foster collaboration.  
 
"These new network data sets are rich and complex," Smyth said. "The structures of digital social networks are changing constantly over time. Our goal in this research project is to develop novel computational techniques to help us better understand and predict the characteristics of these large networks."  
 
Center researchers study computer algorithms that can harness the vast amounts of digital data available in the 21st century. Then they apply them intelligently to solve real-world problems, such as the development of personalized software that can interpret and prioritize e-mails without human intervention.  
 
The center's research spans topics including Web search engines, statistical text mining, image and video data analysis, ocean and atmospheric sciences, analysis of biological and genomic data, sensor networks and medical diagnosis.  
 
The multidisciplinary research effort, which spans the spectrum from theoretical computer science to the social sciences, will involve faculty members with expertise ranging from algorithms to graph visualization, machine learning, data mining, statistics, sociology and behavioral science.  
 
Funding was awarded through the U.S. Department of Defense's Multidisciplinary University Research Initiative. The program supports basic research involving more than one science and engineering discipline and seeks to apply discoveries to both commercial and military uses.  
 
--Jason Mednick, University Communications, 949-824-5951  
--Sherry Main, Donald Bren School of Information and Computer Sciences, 949-824-1562  
 
View campus press release