A Concept-Based Model for Information Retrieval From Holy Quran

Searching for words and concepts in the Holy Quran is facing many challenges due to Arabic language, the specific literature and different writing compared to other Arabic books, and requires a different approach to text mining and information retrieval. In this paper, a new model based on the concept mining approach is proposed to effectively retrieve information from the Holy Quran. In the proposed model, using one of the deep learning algorithms called wrod2vec, while mapping words into concepts, a vector is created for each verse of the Quran representing the concepts mentioned in each verse, and the searched word is converted into the vector in the same space to retrieve the most closely related verses to the searched vector. The results of the experiments indicate a much higher performance of the proposed retrieval model compared to other similar models.