A novel dataset for quranic words identification and authentication

Abstract

Quran is the holy book for Muslims around the world. For the past fourteen centuries after its revelation, ithas been preserved in all possible ways from any distortions. The huge increase in Internet usage and the spread of digital media lead to the development of many websites, services, and applications related to Quran. These efforts include the conversion of Quranic verses, translations, explanations,tafseer and other Quranic sciences into digital formats. Some of these efforts are foundless authentic. The authentication dependson correct identification of Quranic words in the text. In this paper, we introduce a novel dataset for Quranic words identification and authentication. The proposed dataset contains more than 93,000 samples with64 features for each extracted in numerical form.The validation tests of the proposed dataset resulted high accuracy average.