A Classification of Al-hur Arabic Poetry and Classical Arabic Poetry by Using Support Vector Machine, Naïve Bayes, and Linear Support Vector Classification

Abstract

Most of the world languages have made strides in analyzing and classifying texts electronically; hence, the use of electronic text has become a great alternative to manual classification as it reduces time, cost, and difficulty. However, in the Arabic language, electronic analysis has not progressed due to several limitations faced by researchers in this field, such as the complexity of the Arabic language, the lack of related research, as well as the use of the classical Arabic language. In addition, Arabic poetry has other limitations, such as the use of a system that uses a single activation function. In this research, a new method was developed for the classification of the classical Arabic poetry and Al-hur poetry. This new approach is based on features that indicate the type of poetry. Pre-processing of some data is important in this new approach as it helps increase the accuracy of classification.