Academic Journals Database
Disseminating quality controlled scientific knowledge

Performance Improvement of ANN Classifiers using PSO

ADD TO MY LIST
 
Author(s): Jayshri D.Dhande | D.R.Dandekar | S.L.Badjate

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: ncipet;
Issue: 7;
Date: 2012;
Original page

Keywords: Artificial Neural Network | BPNN | RBFNN | SVM | and Particle swarm Optimization (PSO)

ABSTRACT
Data classification has been studied widely in the field of Artificial Intelligence, Machine Learning, data mining, and pattern recognition. Up to the present, the development of classification has made great achievements, and many kinds of classified technology and theory will continue to emerge. The aim of this paper is two fold. First, we present the experimental study of different Artificial Neural Networks classifiers for classification of radar returns from Ionosphere dataset and Bupa liver disorder dataset. Second, we propose a novel classification system based on particle swarm optimization (PSO) to improve the generalization performance of the SVM classifier. For this purpose, we have optimized the kernel parameters of SVM classifier. The experiments were conducted on Jonhs Hopkins Ionosphere dataset and Bupa Liver Disorder dataset. The comparison of different Neural Network classifiers and PSO-SVM is done based on Ionosphere dataset and Bupa Liver Disorder dataset from UCI machine learning repository. The results show that RBFNN typically provide better classification results. When comparing to techniques applied to binary classification problems. Also SVM Classifier with RBF kernel gives best classification accuracy on training set. And PSO-SVM classifier with optimized kernel parameter selection for classification of radar returns from ionosphere dataset and Bupa Liver Disorder gives better accuracy and improves the generalization performance.
Why do you need a reservation system?      Affiliate Program