Improving performance of a multiple classifier system using self-generating neural networks

Authors:
Hirotaka Inoue;Hiroyuki Narihisa
Affiliations:
Department of Electrical Engineering and Information Science, Kure National College of Technology, Kure-shi, Hiroshima, Japan;Department of Information and Computer Engineering, Okayama University of Science, Okayama-shi, Okayama, Japan
Venue:
MCS'03 Proceedings of the 4th international conference on Multiple classifier systems
Year:
2003

Citing 10
Cited 0

C4.5: programs for machine learning

C4.5: programs for machine learning
Bagging predictors

Machine Learning
Self-organizing maps

Self-organizing maps
Data mining: concepts and techniques

Data mining: concepts and techniques
Soft Margins for AdaBoost

Machine Learning
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Forward and Backward Selection in Regression Hybrid Network

MCS '02 Proceedings of the Third International Workshop on Multiple Classifier Systems
Optimizing a Multiple Classifier System

PRICAI '02 Proceedings of the 7th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Bagging, boosting, and C4.S

AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, multiple classifier systems (MCS) have been used for practical applications to improve classification accuracy. Self-generating neural networks (SGNN) are one of the suitable base-classifiers for MCS because of their simple setting and fast learning. However, the computational cost of the MCS increases in proportion to the number of SGNN. In an earlier paper, we proposed a pruning method for the structure of the SGNN in the MCS to reduce the computational cost. In this paper, we propose a novel pruning method for effective processing. The pruning method is constructed from an on-line pruning method and an off-line pruning method. We implement the pruned MCS with two sampling methods. Experiments have been conducted to compare the pruned MCS with the unpruned MCS, the MCS based on C4.5, and k- nearest neighbor method. The results show that the pruned MCS can improve its classification accuracy as well as reducing the computational cost.