This document presents a study on the classification of modern Arabic poetry using machine learning techniques, specifically focusing on four categories: love, Islamic, social, and political poems. It discusses the challenges posed by the complexity of the Arabic language and introduces methods such as Naïve Bayes, Support Vector Machines, and Linear Support Vector Classification, emphasizing the importance of data preprocessing to enhance classification accuracy. The research highlights the lack of available datasets and the methodology adopted for effective classification in the context of Arabic poetry.