Validating a New Collective Intelligence Technology for Accurate Ranking Using Artificial Swarm Intelligence

We introduce hyperswarm ranking, a new collective intelligence technology based on the biological principle of Swarm Intelligence, and show that it enables networked human groups to collaboratively rank independent sets of items with significantly higher accuracy than traditional methods. While prior approaches, such as survey-based Wisdom of Crowd (WoC) techniques, are effective at amplifying groupwise accuracy, we show that this new approach significantly outperforms on a series of general knowledge questions, producing rankings that are 8.1% more accurate than WoC (p < 0.01). This translates into an impressive 39.5% amplification of the traditional “WoC effect.” Finally, we show that the use of hyperswarm ranking enables networked human groups to generate groupwise rankings much faster than other commercially available tools that leverage the accuracy benefits of Artificial Swarm Intelligence, cutting the time required of human participants by more than half.