0

I have a set of n samples, each associated with a set of x observations (real numbers) and y probabilities of outcomes (0 to 1 for each). I want to find subsets of observations that predict particular outcomes, where the number of subsets that predict a particular outcome is unknown.

For example, an example desired answer would be 'high values of these predictors is associated with high probability of this particular outcome.'

This seems to be some sort of mixture model clustering problem, but I'm not terribly familiar with machine learning algorithms, so thought I would see what this community thinks. This may be a question for stackoverflow instead, I'm not sure.

flag
I would agree that this is not appropriate for this forum (and I would guess it would be too vague for stack overflow as well). – Igor Rivin Nov 7 2011 at 20:53
1 
 stats.stackexchange.com might be able to help – David Roberts Nov 7 2011 at 21:09
3 
 To get the best answer out of a stackexchange site, the question needs to be more focussed, like "I need an intro text to machine learning that handles this case...", and you give some more detail (quantities for n,x, and y might be useful). Otherwise, you might find a blog or newsgroup or something that might handle discussing your situation and speculating on a solution for it. Gerhard "Watch Another Year Go By" Paseman, 2011.11.07 – Gerhard Paseman Nov 7 2011 at 21:30
"Watch Another Year Go By"??? – Igor Rivin Nov 7 2011 at 21:37

closed as off topic by Igor Rivin, David Roberts, Andy Putman, Ryan Budney, Daniel Moskovich Nov 7 2011 at 23:35

Browse other questions tagged or ask your own question.