i don't know how to do this in an algorithm...
lets simplify the problem
suppose you have a 2d array of points - excuse me while i make one
how do we algorithmicly determine the fact that a larger amount of points are in the top right corner?
my math knowledge isn't great enough yet - can we use some kind of regression or even just simple averaging? suppose you were to calculate the average position of the first three points in the plane. then, as you add in the other points, recalculate the averages. if it skews the average too much, assume that brick isn't part of the group
barnabas, to my knowledge, this is what they call a 'hard' computer problem. humans can look at the graph and say pretty confidently that most of the points are in the top right corner - computers are not so lucky.