R help - Apr 2014 - Finding best cut points to discretize continuous values for ID3 algorithm

Hi all,

I am a senior student of MIS department at Bogazici University, Istanbul.
My graduation project's focus is on data mining using R.

I would like to write a function which takes find best(or near to best) cut
points to discretize continuous values to minimize the conditional entropy
based on a class variable.

For instance,

a= c(1,2,3,4,5,6)
b= c(a,a,b,b,c,c)

The function I would like to write is supposed to tell me that cut points
should be around 2.5 and 4.5 to minimize the entropy which in this case
would be zero because I would have perfect information.

I've searched for a similar function but I didn't manage to find one. I
wonder if there is any function like this already? If no, would it be
helpful if I create a function like this?

Kind regards,
Furkan Gursoy
Senior Student at MIS
Bogazici University '14

	[[alternative HTML version deleted]]