Here is a link to the ID3 algorithm by
Ross Quinlan:
http://en.wikipedia.org/wiki/ID3_algorithm
The algorithm iteratively divides the
elements of S based on all attribute values of all objects, which are said
elements of S. So it picks the group of features and weights that make the
most “efficient” split. That separates S into S0 and S1. Its use of the “entropy”
analogy to evaluate efficiency of the split seems a bit of anthropopathy.
Has anyone used ID3 to identify classes in
clustering analysis?
-Rich
Sincerely,
Rich Cooper
EnglishLogicKernel.com
Rich AT EnglishLogicKernel DOT com
9 4 9 \ 5 2 5 - 5 7 1 2