Big data continues to grow and so does the potential to mine information, analyze it and connect dots some might not know existed.
As someone who was intrigued by such things, David W. Allen — a Los Alamos National Lab veteran turned Boulder entrepreneur — created a meetup group thinking a few others might want to talk shop about graphs and network analysis.
“The amount of data that we’re generating these days is unbelievable,” he said. “Having these kinds of groups allows us to tackle what data is important, how do you handle it and what’s even possible.”
When Allen launched the Graph Nerds of Boulder meetup last year, he anticipated that the first meeting might garner two, maybe three attendees.
Thirteen people showed up to the first meetup.
Each meeting since has grown in size and attracted a wide variety of people — not simply network analysts.
“It’s a very niche topic, but I think it’s also really accessible,” he said, adding that “everyone’s welcome.”
This Wednesday, Graph Nerds’ sixth meetup will consist of a presentation from Sears Merritt, a Ph.D. student in the University of Colorado-Boulder’s computer science department:
In many online social systems, interactions between individuals are observable but their friendships are hidden. Using behavioral data from 18 billion interactions among 17 million individuals and an anonymous online survey, we develop highly accurate predictors that automatically extract latent friendships between individuals with over 95% accuracy. Using these predictors we infer the hidden social network of individuals in the online game Halo: Reach and find that its structure is recognizably different from the popular online social network, Facebook. These results clarify the nature of friendship in online social environments and suggest new opportunities and new privacy concerns for friendship-aware applications that do not require the disclosure of private friendship information.
For more information on the meetup group and to register for the meeting, click here.