17 Mar 2004
CS 3243 - Learning
37
Information gain
nFor the training set, p = n = 6, I(6/12, 6/12) = 1 bit
n
nConsider the attributes Patrons and Type (and others too):
n
n
n
n
n
nPatrons has the highest IG of all attributes and so is chosen by the DTL algorithm as the root