There are 21 judges whose probabilty for making correct
decision(binary decision of accept or reject) is all 0.7. They will
aggregate their decision with the simple majority rule(more than half
the number wins the decision). What is their probability for correct
decision?

   pr=:21#0.7
   len=:#pr
   cor=:|: (#~ (>.2%~len) <: +/"1) #: i. 2^len
   wro=:-. cor
   +/ */ (pr* cor) + (1-pr)* wro
0.97361

Suprisingly, the probability goes quite high than their original
correctness probability.

Now, play it with five judges: 0.95, 0.95, 0.90, 0.90, 0.80

   pr=:0.95 0.95 0.90 0.90 0.80
   len=:#pr
   cor=:|: (#~ (>.2%~len) <: +/"1) #: i. 2^len
   wro=:-. cor
   +/ */ (pr* cor) + (1-pr)* wro
0.992945

1) Could you come up with more elegant, efficient solutions?


2) What would happen if the least correct judge(0.80) from the second
case just follows the first judge(0.95)'s decision?


3) What is the condition that the result probabilty is higher than the
best person's correct decision probability?
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to