Data classification using Support vector Machine (SVM), a simplified approaCH

Abstract

In all our day to day activities we will be classif ying things based on situations and on our needs. H uman beings do classification of any kind by their natural percept ion. Classifying data is a common task in machine le arning which requires artificial intelligence. Support vector Machine (SV M) is a new technique suitable for binary classific ation tasks. SVMs are a set of supervised learning methods used for classificat ion, regression and outliers detection. The SVM cl assifiers work for both linear and nonlinear class of data through Kernel t ricks. A Support Vector Machine is a discriminative classifier formally defined by a separating hyperplane. In other words, given l abeled training data, the algorithm outputs an opti mal hyperplane which categorizes new samples. In this paper, use of SVM for data classification is presented in a simplifie d way. Discussions are justified with illustrative practical examples. An effective algorithm is developed for data classific ation on python platform using sklearn tool kit. The results are exhibited both sy mbolically and graphically. This paper is expected to be an insight for desired readers and researchers in implementing their ideas of item classification using SVM.