| | |
Summary: International Journal of Pattern Recognition
and Artificial Intelligence
Vol. 19, No. 3 (2005) 323353
c World Scientific Publishing Company
LINEAR DISCRIMINANT TREES
OLCAY TANER YILDIZ and ETHEM ALPAYDIN
Department of Computer Engineering
Bogazi¸ci University
34342 Istanbul, Turkey
yildizol@cmpe.boun.edu.tr
alpaydin@boun.edu.tr
We discuss and test empirically the effects of six dimensions along which existing decision
tree induction algorithms differ. These are: Node type (univariate versus multivariate),
branching factor (two or more), grouping of classes into two if the tree is binary, error
(impurity) measure, and the methods for minimization to find the best split vector and
threshold. We then propose a new decision tree induction method that we name linear
discriminant trees (LDT) which uses the best combination of these criteria in terms of
accuracy, simplicity and learning time. This tree induction method can be univariate
or multivariate. The method has a supervised outer optimization layer for converting a
K > 2-class problem into a sequence of two-class problems and each two-class problem
|
