• search hit 1 of 17
Back to Result List

Computing Generic Causes of Revelation of the Quranic Verses Using Machine Learning Techniques

Please always quote using this URN: urn:nbn:de:bvb:20-opus-66083
  • Because many verses of the holy Quran are similar, there is high probability that, similar verses addressing same issues share same generic causes of revelation. In this study, machine learning techniques have been employed in order to automatically derive causes of revelation of Quranic verses. The derivation of the causes of revelation is viewed as a classification problem. Initially the categories are based on the verses with known causes of revelation, and the testing set consists of the remaining verses. Based on a computed thresholdBecause many verses of the holy Quran are similar, there is high probability that, similar verses addressing same issues share same generic causes of revelation. In this study, machine learning techniques have been employed in order to automatically derive causes of revelation of Quranic verses. The derivation of the causes of revelation is viewed as a classification problem. Initially the categories are based on the verses with known causes of revelation, and the testing set consists of the remaining verses. Based on a computed threshold value, a naïve Bayesian classifier is used to categorize some verses. After that, using a decision tree classifier the remaining uncategorized verses are separated into verses that contain indicators (resultative connectors, causative expressions…), and those that do not. As for those verses having indicators, each one is segmented into its constituent clauses by identification of the linking indicators. Then a dominant clause is extracted and considered either as the cause of revelation, or post-processed by adding or subtracting some terms to form a causal clause that constitutes the cause of revelation. Concerning remaining unclassified verses without indicators, a naive Bayesian classifier is again used to assign each one of them to one of the existing classes based on features and topics similarity. As for verses that could not be classified so far, manual classification was made by considering each verse as a category on its own. The result obtained in this study is encouraging, and shows that automatic derivation of Quranic verses’ generic causes of revelation is achievable, and reasonably reliable for understanding and implementing the teachings of the Quran.show moreshow less

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Metadaten
Author: Mohamadou Nassourou
URN:urn:nbn:de:bvb:20-opus-66083
Document Type:Preprint
Faculties:Philosophische Fakultät II (bis Sept. 2007) / Institut für deutsche Philologie
Language:English
Year of Completion:2011
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik
GND Keyword:Text Mining; Koran
Tag:Causes of revelation; Quran; Statistical classifiers; Text mining; Text segmentation
Release Date:2011/11/17
Licence (German):License LogoDeutsches Urheberrecht