Enhancing the Bert Model for Qur'anic Question Answering System

Machine Reading Comprehension is a rapidly evolving field that aims to bridge the gap between human-like understanding and machine text processing. With advancements in deep learning and the development of sophisticated models, MRC systems are becoming increasingly capable of understanding context and providing accurate answers. Driven by the increasing demand for advanced question-answering (QA) systems on the Holy Qur'an, the inaugural Qur'an Question Answering shared task (Qur'an QA 2022 OSACT 2022), the second edition in Arabic NLP 2023. The Qur'an QA 2023 encompasses two essential sub-tasks: Passage Retrieval (PR) and Machine Reading Comprehension (MRC). The shared task aims to enhance the capabilities of QA systems and facilitate deeper engagement with the Qur'an. This research paper focuses on developing an improved BERT model, implemented using the KNIME platform, and tailored for Arabic. We aim to utilize this model for Qur'anic Question Answering, resulting in an Area under the Curve (AUC) of 0.5 and an accuracy rate of 0.997.