Paper Details: Downloads: 441
Serial Number: P1120847488
Title: PERFORMANCE EVALUATION OF SVM KERNELS USING HYBRID PSO-SVM
Authors: S. Sivakumari and R. Praveena Priyadarsini and P. Amudha
Abstract: Abstract This paper presents a hybrid data mining approach for knowledge extraction and classification in databases. Support Vector Machine (SVM) classification is an active research area which solves classification problems on different domains. Particle Swarm Optimization (PSO) is a new evolutionary computation technique in which each potential solution is seen as a particle with a certain velocity flying through the problem space. This study combines support vector machines with particle swarm optimization for classification. The idea is to classify the datasets using support vector machines with various kernels and optimize the classifier using particle swarm optimization. SVM classifier with 10-fold cross validation is applied inorder to validate and evaluate the provided solutions. The approach has been implemented and tested on benchmark datasets of various sizes. Experiment results show that the hybrid PSO-SVM approach finds interesting patterns and provides improved classification performance.
Keywords: Data mining, Support Vector Machines, Particle Swarm Optimization, Classification, Knowledge Extraction, Cross validation.
Journal/Conference: International Journal of Artificial Intelligence and Machine Learning
Volume: 9
Issue: 1
Submission Date: 11/19/2008 12:00:00 AM
Review Date: 1/5/2009 12:00:00 AM
Publishing Date: 2/5/2009 12:00:00 AM
Article Downloads: 441
Download:

Facebook