Comparing classifiers and metaclassifiers

Authors:
Elio Lozano;Edgar Acuña
Affiliations:
University of Puerto Rico at Bayamón, Computer Science Department, Bayamón, Puerto Rico;University of Puerto Rico at Mayagüez, Mathematics Department, Mayagüez, Puerto Rico
Venue:
ICDM'11 Proceedings of the 11th international conference on Advances in data mining: applications and theoretical aspects
Year:
2011

Citing 9
Cited 1

C4.5: programs for machine learning

C4.5: programs for machine learning
Ensembles of Learning Machines

WIRN VIETRI 2002 Proceedings of the 13th Italian Workshop on Neural Nets-Revised Papers
Experiments with Classifier Combining Rules

MCS '00 Proceedings of the First International Workshop on Multiple Classifier Systems
Ensemble Methods in Machine Learning

MCS '00 Proceedings of the First International Workshop on Multiple Classifier Systems
Methods for Designing Multiple Classifier Systems

MCS '01 Proceedings of the Second International Workshop on Multiple Classifier Systems
Classifier ensembles: Select real-world applications

Information Fusion
Ensemble of single-layered complex-valued neural networks for classification tasks

Neurocomputing
A data driven ensemble classifier for credit scoring analysis

Expert Systems with Applications: An International Journal
Combination of multiple classifiers for post-placement quality inspection of components: A comparative study

Information Fusion

Ensemble methods for advanced skier days prediction

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

A metaclassifier is a technique that integrates multiple base classifiers. In this paper a hybrid meta-classifier algorithm based on generative and non-generative methods is proposed. Five well-know strong classifiers are used for the non-generative method and bagging was used for generative method. The performances of the five base classifiers, their ensembles based on bagging, and the proposed hybrid metaclassifier are compared using classification error rates. Eight different datasets coming from the UCI Machine Learning database repository are used in the experiments.