Juz Classification of English-Translated Qur'anic Verses Using Support Vector Machine

While numerous studies have focused on the classification of Qur'anic verses based on thematic content, limited attention has been given to classification based on juz. This study proposes an automated classification model to assign English-translated Qur'anic verses to their corresponding juz using the Support Vector Machine (SVM) algorithm. The input consists of verse-level translations, while the output is the predicted juz label. As traditional juz classification is typically performed manually and subjectively, this research introduces a computational approach that is both objective and scalable. The dataset comprises the complete English translation of the Qur'an, totaling 6,236 verses annotated with juz labels. The experimental procedure consists of several stages, including data preprocessing, TF-IDF-based feature extraction, and model implementation using Support Vector Machine, Naive Bayes, and K-Nearest Neighbors. To address the issue of data imbalance, an oversampling technique is employed. Experimental findings indicate that the SVM model achieves the highest classification performance, with an accuracy of 84%, macro precision of 82%, macro recall of 84%, and macro F1-score of 83%.