Many people in today’s culture are familiar with social media and use various platforms to network with one another. Few people, however, really understand the complexity of social media networks like Itai Himelboim.

A majority of Himelboim’s research examines social media networks through Twitter, Facebook and other platforms to identify unique patterns of social interactions, key influential users and emerging communities, and to track information flow.

“Almost anything you do on social media, you display engagement with people, and network analysis examines the structures of these patterns of engagement,” Himelboim said. “If you engage by following someone, information flows to you. If you engage by retweeting, information flows through you to others. And, that’s really what I believe is the core of understanding social media activity: patterns of social interactions. Network analysis puts the ‘social’ back in social media.

“When we interact with others on social media, we open channels for information flow, and therefore determine where we receive information from,” Himelboim also said. “By mapping how people interact, we are tracking content flows, and more importantly, the boundaries of information flow.”

Himelboim’s research focuses on information flows among individuals and between organizations and the public, primarily in conversations about politics, international affairs and commercial brands.

“Studying the networks formed by elections-related conversations, we clearly see the information silos we all create: we primarily interact with like-minded others and consume information from like-minded news sources,” he said. “This is far from the marketplace of opinions and ideas that was once expected of the Internet.”

Himelboim, who came to the U.S. from Israel as a Ph.D. student, got interested in network analysis while he worked on projects with a computer science and molecular biology colleague. In those early days of social media, they gathered data from Newsgroups’ discussion boards. These projects led him to an internship with Microsoft Research.

“That’s what really got us started,” Himelboim said. “Using network analysis, we mapped the relationship and patterns of information flow among individuals talking about politics. A small minority of people were found to be located in unique positions in the network, attracting large and disproportionate number of replies. When most people posted content, almost nothing happened, but when these few discussion catalysts posted, a conversation was evoked. They played a key role in terms of information flow and setting the agenda for political discourse.”

In addition to his research, Himelboim is motivated by teaching students how to understand big social media data, analyze it, report it, find the stories and engage in real time.

“We need to remember that our graduates today are expected to know about social media, but just because you use it, doesn’t mean you understand it,” he said.

His newest project is designing the new SEE Suite at Grady College—a cutting-edge social media monitoring space—where students can examine big, cross-platform, social media data; use a variety of software to analyze it and identify key actionable insights; interact with agencies via teleconferencing; and present their work.