Outfitted with the right algorithms and analytics tools, you should essentially be able to measure the social behavior of the human race. You should be able to determine and predict how and why humans interact with one another, how they share information, and how they influence one another, said Jure Leskovec, an assistant professor of computer science at Stanford.
After all, he says, "Everything we do, we do on the Web." Leskovec shared his advanced analytical thinking and early research results during a Webinar, "Mining Online Data Across Social Networks: Capturing Data, Modeling Patterns, Predicting Behavior," hosted by Stanford yesterday. As the Webinar title suggests, Leskovec concentrates his research on mining and modeling of data from large social and information networks and the study of how influence and information spreads through these online connections.
"The Web is the laboratory in which we can study the pulse of humanity," noted Leskovec, who also is a member of Stanford's InfoLab and Artificial Intelligence Lab communities, during the Webinar.
Leskovec said his research already is proving useful, not only in predicting the most influential nodes or individuals in any network, but also in identifying where information may best be placed to ensure that it spreads virally throughout a network. The research also paves the way for developing metrics for measuring social interaction, as well as determining how and why social networks evolve in the way they do, Leskovec added.
During the presentation, Leskovec said he believed his research would be important to fields ranging from marketing to Web development because it will make possible the use of these metrics for predicting how best to spread any message, and which online networks survive and which do not.
As a practitioner of online community building on a number of sites and now the person primarily responsible for building the community here at AllAnalytics.com, I easily filled in the blanks as I listened. How prescient researchers would have seemed a few years ago if they had been able to predict the collapse of then-social media giant MySpace and the rise of its upstart competitor Facebook as world leader, or at least leader of the online world, by using sophisticated analytics that vastly improve data mining.
And imagine the value in social media marketing, one of the most rapidly growing fields on the Internet, if such analytics capabilities could predict which potential online connections held the greatest influence within targeted communities, and which are most likely to spread content shared with them.
If the future of social media interests you, be sure to join us for an instant e-chat tomorrow, August 18, at 2 p.m. Eastern. (If you haven't yet registered on AllAnalytics.com, please fill out a quick form here to participate in the chat.) We'll be talking with AllAnalytics.com bloggers Joe Stanganelli and Pierre DeBois, both social and Web analytics experts, about what analytics may be able to tell us about the social media world. We hope you'll join in the conversation.
Meanwhile, what do you think can be learned from metrics that measure interaction on social media and the Web, and how might they affect your company? Please leave a comment and start a discussion below.