The Wisdom of Crowds II: Cognition

Lecture on 11.10.2007

Presentation

Cognition problems are problems that involve finding a definite answer. Under certain circumstances, groups perform extremely well in cognition problems; in others, not so well. During this class period, we looked at several different types of cognition problems and some methods to take advantage of group knowledge.

The students were asked to individually guess how many liters of water were in the Baltic Sea. The average answer was orders of magnitude off-target, although a couple individual answers were closer. This was most likely because of the corollary of the Condorcet Jury Theorem, because we as a group of computer scientists have a very low percentage individually of being right. If we had been more people, there probably would have been an even less chance of being right.

• People put in effort voluntarily
• Only people interested in the topic participate (helps to reach the >50% mentioned in the Condorcet Jury Theorem)
• Reaction on market changes is possible
• It's not about the personal opinion but about “What does the majority think?”

• Topics for upcoming examinations
• When will the train conductor strike end?

“I arrive at this conclusion in a somewhat roundabout way. I have long been fascinated with systems that tap into the “wisdom of crowds” – systems that, in fact, have much in common with Darwinian evolution. Such systems doubtfully conflict with anyone's religion, and yet, I see the same sort of resistance to them as I see to evolution. The arguments against them are remarkably similar.”
(Rob Brown, http://karmatics.com/docs/evolution-and-wisdom-of-crowds.html

• Are prediciton markets used e.g. by the big market researchers like Gartner?

• Google Group Prediction Markets
• Yahoo Event about Prediction Markets