
FlickR co-founder, Caterina Fake recently rolled out Hunch, an online decision-making tool. It's sort of like the old "Magic Eight Ball", but now with collective intelligence to optimize a sort of decision making wizard.
But wait, there's more. In addition to user-generated question refinements (question/response "training"), it also uses statistical inferences associated with the discipline of machine learning.
It's all The math is beyond me, but key highlights from Hunch's "How Hunch Works" page makes for an interesting read:
- In choosing what to ask you, Hunch's question selection algorithm tries to do two things. First, it tries to find a question which will discriminate well among the remaining possible decision outcomes for you - thus filtering the remaining choices from "many" to "fewer". Second, the algorithm looks for a question which can help optimize and rank the remaining decision results to present you with the ones you'll like the most.
- As you answer questions, Hunch can narrow down your possible decision outcomes because each outcome can be "trained" to correspond with each question's answers.
- Any logged in user can set initial training or correct existing training, in addition to proposing new topics, questions to ask, and decision outcomes.
- When a user clicks "Yes" or "No" to indicate whether or not they like a decision result, Hunch incrementally strengthens or weakens the mathematical correlation between that result and any 'Teach Hunch About You' questions that have been answered so far.