From the Wires
Big Data Algorithm Secrets Revealed by Experts at Rocket Fuel's Successful Chicago Event
Fast-Growing Company Hosted a Lively Discussion That Included Author Christopher Steiner, Obama for America's Chief Data Scientist Rayid Ghani, as Well as Other Leading Experts in the Field
By: Marketwired .
Dec. 18, 2012 12:41 PM
REDWOOD CITY, CA -- (Marketwire) -- 12/18/12 -- Rocket Fuel, the leading provider of artificial intelligence advertising solutions for digital marketers, recently hosted its successful "Advertising That Learns: Where Artificial Intelligence Met Madison Avenue" on December 5, 2012. Leaders from the interconnected ecosystem of digital advertising, politics, academia, and industry met to discuss the growing impact of artificial intelligence and big data algorithms on our daily lives.
Speakers at the event included:
The event, which was held at Chicago's Mid America Club, opened with a keynote by Steiner, and followed with a panel discussion featuring the headline speakers. George John, the founder and CEO of Rocket Fuel, led the conversation and posed key questions to the panelists about how artificial intelligence (AI) and big data algorithms have impacted everything from politics, to education, commerce, medicine, advertising, and beyond.
As Steiner pointed out up front, algorithms are a set of instructions that tell a machine what to do with a piece of information. "But algorithms have grown to be incredibly complex and nimble," he said. "They're no longer rote sequence sets of instructions. They can be dynamic, they can learn, they can adapt, they can evolve. And that's of course where AI comes in. Algorithms in fact have evolved to the point where we're not always the ones shaping them; they shape themselves."
Steiner's closing line, "The guy with the data is going to win," started an interesting conversation. As Tobaccowala sees it, having the data is just the beginning of the battle. "If you don't have the data, it's unlikely that you're going to win," he said. "On the other hand, if you have all of the data, that's not sufficient to win."
Phillips agreed. "The marketer's dilemma is knowing that there are a number of different factors we're looking at that will tell us who our ideal and most profitable customer is. But how do we parlay that into the marketing side of things using similar technologies to really hone in on the information? It's easy to reach a lot of people, but ideally we want to reach the right people that are going to fit within our optimal pricing model -- which will yield better profitability."
Not surprisingly, getting to profitability is a key concern for all of the speakers. As Grossman said, "When you bring together best-practice analytic operations, lifecycle management for the algorithms, access to the data, understanding latency with the analytic infrastructure and the algorithms, I think of that as the perfect storm [for big data]. And the communities that best understand how to exploit that have a huge competitive advantage."
Though they came from different industries, with different challenges, the panelists all agreed with Tobaccowala's statement: "Clearly, every large company has discovered math is helping them." But another key point he made was "One of the things I've said is where the world might be digital, people remain analog. You actually can inspire people through emotion, but they remain analog because they're fearful of change."
This fear of change is everywhere, and, people's biggest fear is being replaced by machines. As Tobaccowala said, "If you hear a little voice inside saying 'your job is obsolete,' believe it. It is... Management's dream is to replace carbon (humans) with silicon (computers)."
John gently countered, saying "It's not so much silicon versus carbon. I like our employees... To me, silicon is just one way of operationalizing in an efficient way."
Ghani agreed, and made a clear point about the critical need for both carbon and silicon when it came to the success of Obama for America 2012. "We had a really good engineering team and analysts who over the past several years had built an infrastructure to put all the data together... We could've made the perfect printout list. But if we had nobody to make the phone calls (or knock on doors), it would've been completely useless. So yes, we had a smart group of people who built models and predicted all these things. That was really important, but it wasn't sufficient. We had both a data infrastructure and an execution infrastructure. Without those things, we couldn't have done anything."
An archive of the event's web simulcast -- viewed by hundreds of people in six countries worldwide -- can be viewed here.
About Rocket Fuel Inc.:
Latest Cloud Developer Stories
Subscribe to the World's Most Powerful Newsletters
Subscribe to Our Rss Feeds & Get Your SYS-CON News Live!
SYS-CON Featured Whitepapers
Most Read This Week