This blog is based on this paper: “Crowds vs Swarms, a Comparison of Intelligence” by Louis Rosenberg, David Baltaxe, and Niccolo Pescetelli.
Recently, I went for a conference organized by Daniel Palmer and Marc Kirschenbaum of John Carroll University on Blended Intelligence. I thought it appropriate to talk about one of the talks. How do we get intelligence from a crowd of people, surveys, interviews? How does nature get intelligence from its beings? Authors claim nature does not aggregate independent samples but works on a closed real-time loop with continuous feedback. Hence, can we have a human swarm similar to a flock of birds or a school of fish and does it result in better intelligence? That is exactly what the authors put to test with their software UNU. UNU works by having a group of knowledgable individuals about a specific topic to come together virtually and decided on an answer for a given question. Each user has a magnet which he/she can use to pull the puck toward their desired answer.
What of the results? Check this article on how it predicted the Kentucky Derby, or read their paper on its prediction for the 2016 Super Bowl; a human swarm of 20 people outperformed (68% correctly) a crowd of 469 football fans (47% correctly). If this doesn’t impress you, well the swarm outperformed 98% of independent individuals in the study. Now, could this be a reason to pool our intelligence in order to tackle more challenging questions facing us in the future? Could this help in finding solutions to climate change that is affecting us more every day.
Do you want to try it? All you need is to sign up, verify your email, and you’ll be in your way to create you first UNU human swarm, or you can just enter one of their open UNUs. Finally check out their tutorial: https://youtu.be/TkAoRUHs5F0