<![CDATA[Arabic has a complex morphological system, particularly in the formation of fi’il (verbs), which in sharaf are classified into fi’il mujarrad and fi’il mazid. Fi’il mujarrad refers to basic verbs without additional letters, while fi’il mazid involves added letters that form specific wazan (patterns). The main problem addressed in this study is the similarity of morphological patterns between fi’il and non-fi’il, which affects classification accuracy. This study aims to identify fi’il in Surah Al-Baqarah and classify them into mujarrad and mazid, along with their wazan patterns, automatically using a rule-based pattern matching approach. The method applies rules based on Arabic morphological patterns, such as fi’il mudhari’ prefixes, word length, and the presence of additional letters in accordance with sharaf principles.The data consist of Qur’anic text processed through preprocessing stages, including load data and tokenization. The system detected 916 candidate fi’il, of which 468 data points were used for evaluation by comparison with manual annotations using a confusion matrix.The results show that the system achieved an accuracy of 75% for fi’il type classification, with precision, recall, and F1-score of 0.77, 0.75, and 0.75, respectively. For wazan classification, the system achieved an accuracy of 69.23%, with weighted average precision of 0.66, recall of 0.69, and F1-score of 0.65. These findings indicate that the rule-based approach is sufficiently effective in detecting fi’il mujarrad and mazid, although performance for certain wazan patterns remains limited due to structural similarities. Therefore, further development of more specific rules and integration with machine learning methods are recommended to improve system accuracy. ]]>
