Text categorisation in Quran and Hadith: Overcoming the interrelation challenges using machine learning and term weighting

Abstract

Quran and Al-Hadith are interrelated in the sense that both often complement each other in interpreting Islamic teachings. In order to gain comprehension of the Quran in detail, it is vital for a Muslim to refer to Al-Hadith in clarifying ambiguities from the Quran. Al-Hadith offers explanations and lends certainties to the abstract concepts depicted in the Quran. With that, this research proposes a method using text categorisation to classify selected categories by determining the interrelation between the resources. Several interrelated cases were simulated by using a combination of different Islamic resources datasets comprising Quran and Hadiths. The selected three categories; Hajj, Prayer, and Zakat, were compared using three classification methods (Naïve Bayes (NB), Support Vector Machine (SVM), and K-Nearest Neighbour (KNN)) via term weighting; Term Frequency – Inverse Document Frequency (TF-IDF). As a result, the SVM, regardless of being used alone or with term weighting, successfully addressed the interrelationship for single- and multi-label classifications. Additionally, SVM attained better accuracy with 10%–20% improvement, when compared to the other methods that had managed to exhibit slight improvement accuracy wise.