Using an ensemble classifier for machine learning applications

  • Authors:

  • Costas Tsatsoulis;Danico Lee

  • Affiliations:

  • Department of Electrical Engineering and Computer Science, Information and Telecommunication Technology Center, The University of Kansas, Lawrence, KS;Department of Electrical Engineering and Computer Science, Information and Telecommunication Technology Center, The University of Kansas, Lawrence, KS

  • Venue:
  • ICCOMP'05 Proceedings of the 9th WSEAS International Conference on Computers
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

We describe an architecture for integrating classification algorithms that have been created by a variety of machine learning methods, trained on the same data set. The ensemble classifier unifies all these classifiers into a single module and uses voting and a reward/punishment system to select the best classifier for a specific data set. In this paper we discuss the theory behind the ensemble architecture, and present its implementation and a set of experiments using a variety of data sets. Our work shows how the ensemble performs as well or better than the best classifier for a specific data set on most occasions.