What should be minimized in a decision tree?

Authors:
Usama M. Fayyad;Keki B. Irani
Affiliations:
Artificial Intelligence Laboratory, Electrical Engineering & Computer Science Department, The University of Michigan, Ann Arbor, MI;Artificial Intelligence Laboratory, Electrical Engineering & Computer Science Department, The University of Michigan, Ann Arbor, MI
Venue:
AAAI'90 Proceedings of the eighth National conference on Artificial intelligence - Volume 2
Year:
1990

Citing 5
Cited 5

Knowledge extraction through learning from examples

Machine learning: a guide to current research
Occam's razor

Information Processing Letters
Induction of decision trees from inconclusive data

Proceedings of the sixth international workshop on Machine learning
Induction of Decision Trees

Machine Learning
Generating production rules from decision trees

IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 1

Controlled Flux Results in Stable Decision Trees

ICTAI '99 Proceedings of the 11th IEEE International Conference on Tools with Artificial Intelligence
Relaxing instance boundaries for the search of splitting points of numerical attributes in classification trees

Information Sciences: an International Journal
Pareto-optimality of oblique decision trees from evolutionary algorithms

Journal of Global Optimization
Generalization behaviour of alkemic decision trees

ILP'05 Proceedings of the 15th international conference on Inductive Logic Programming
Hierarchical linear support vector machine

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we address the issue of evaluating decision trees generated from training examples by a learning algorithm. We give a set of performance measures and show how some of them relate to others. We derive results suggesting that the number of leaves in a decision tree is the important measure to minimize. Minimizing this measure will, in a probabilistic sense, improve performance along the other measures. Notably it is expected to produce trees whose error rates are less likely to exceed some acceptable limit. The motivation for deriving such results is two-fold: 1. To better understand what constitutes a good measure of performance, and 2. To provide guidance when deciding which aspects of a decision tree generation algorithm should be changed in order to improve the quality of the decision trees it generates. The results presented in this paper can be used as a basis for a methodology for formally proving that one decision tree generation algorithm is better than another. This would provide a more satisfactory alternative to the current empirical evaluation method for comparing algorithms.