INTRODUCTION
SVM is a learning model that makes use of a non linear mapping function to transform data in an input space in other to make a problem to be linearly separable [4]. SVM is based on decision plane approach where the training data is mapped into a higher dimensional space and separated by a plane; this plane defines two or more classes of data [10]. In other words SVM performs classification by constructing a hyperplane that optimally separates the data into two categories. SVM automatically discovers the maximum separating hyperplane.
In this project, LIBSVM will be used as a tool for single class classification and multiclass classifications. For the single classification, the “ionosphere dataset” will be used while for the multiclass, “the wine dataset” will be used.
LIBSVM is an integrated enhanced software for support vector classification, ( C-SVC, nu-SVC ), regression (epsilon-SVR, nu-SVR ) and also used for distribution estimation
( one-class SVM ). It supports multi-class classification. The reason why LIBSVM is used is because it posses the following characteristics: different SVM formulation, efficient multiclass classification, cross validation for model selection, probability estimates, weighted SVM for unbalanced data, both C++ and Java sources e.t.c [1].