Poem classification using machine learning approach

Abstract

The collection of poems is ever increasing on the Internet. Therefore, classification of poems is an important task along with their labels. The work in this paper is aimed to find the best classification algorithms among the K-nearest neighbor (KNN), Naïve Bayesian (NB) and Support Vector Machine (SVM) with reduced features. Information Gain Ratio is used for feature selection. The results show that SVM has maximum accuracy (93.25%) using 20% top ranked features.