Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Linear Discriminant Trees Olcay Taner Y ld z yildizol@cmpe.boun.edu.tr

Summary: Linear Discriminant Trees
Olcay Taner Yldz yildizol@cmpe.boun.edu.tr
Ethem Alpaydn alpaydin@boun.edu.tr
Department of Computer Engineering, Bogazici University, 80815 _
Istanbul Turkey
Univariate decision trees at each decision
node consider the value of only one feature
leading to axis-aligned splits. In a multi-
variate decision tree, each decision node di-
vides the input space into two with an ar-
bitrary hyperplane leading to oblique splits.
In this paper, we propose linear discriminant
trees where each node implements Fisher's
linear discriminant analysis (LDA) to make
a binary split. On twenty data sets from
the UCI repository, we compare the linear
discriminant trees with the univariate deci-
sion tree method ID3, multivariate decision
tree methods CART, neural trees and linear


Source: Alpaydın, Ethem - Department of Computer Engineering, Bogaziçi University


Collections: Computer Technologies and Information Sciences