ENHANCING THE PERFORMANCE OF SVM CLASSIFICATION BASED ON FISHER LINEAR DISCRIMINANT

B.Sunitha, M.Seetha

Abstract: Support Vector Machine (SVM) is one of the important classification method used in many areas. Normal support vector machine is not suitable for classification of large data sets due to its high training complexity. Training of SVM with data number n has time complexity between O(n2 ) and O(n3 ). This paper introduces a novel data reduction method Fisher’s decision tree for SVM classification. It is a classifier uses the dimensionality reduction of Fisher Linear Discriminant and decomposition strategy of decision trees. The proposed classifier has distinctive advantages on dealing with large data sets

Keywords: SVM, decision tree, Fisher Linear Discriminant

DOI: https://doi.org/10.15623/ijret.2014.0317009