Using decision tree models and diversity measures in the selection of ensemble classification models

Authors:
Mordechai Gal-Or;Jerrold H. May;William E. Spangler
Affiliations:
School of Business Administration, Duquesne University, Pittsburgh, Pennsylvania;Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, Pennsylvania;School of Business Administration, Duquesne University, Pittsburgh, Pennsylvania
Venue:
MCS'05 Proceedings of the 6th international conference on Multiple Classifier Systems
Year:
2005

Citing 7
Cited 1

Bagging predictors

Machine Learning
Error reduction through learning multiple descriptions

Machine Learning
On the Accuracy of Meta-learning for Scalable Data Mining

Journal of Intelligent Information Systems
On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization

Machine Learning
Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy

Machine Learning
Application of majority voting to pattern recognition: an analysis of its behavior and performance

IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans

Ensembles of ARTMAP-based neural networks: an experimental study

Applied Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a contingency-based approach to ensemble classification. Motivated by a business marketing problem, we explore the use of decision tree models, along with diversity measures and other elements of the task domain, to identify highly-performing ensemble classification models. Working from generated data sets, we found that 1) decision tree models can significantly improve the identification of highly-performing ensembles, and 2) the input parameters for a decision tree are dependent on the characteristics and demands of the decision problem, as well as the objectives of the decision maker.