Arabic Poems Classification According to their Eras and Topics

Abstract

Arabic text classification has faced many difficulties because of the Arabic language natural especially poetry which is very challenging form of the Arabic text. The big growth of the Arabic internet content in the last years has raised up the need for Arabic text processing tools because of its unstructured data. And because poetry classificat ion didn’t get the attention that it deserves, it becomes one of our motivat ions for this work. The lack of open source corpora, lead us to build two corpora ‘Adab’ and ‘Aldiwan’ of Arabic poetry. We used these two corpora to build a system of Arabic Poems Classification according to their category, era and topic. We used machine learning techniques in which we applied a numerous methods and classifiers. The best results were obtained using the LSVC (Linear Support Vector Classification) algorithm. The last model that we obtained in our experiments, we used it to build our application which we called “Saline Classificat ion”.

Keywords: Poetry Classification, Arabic Language, Category Identification, Era Identification, Topic Identification, Machine Learning