Keyword based Clustering Technique for Collections of Hadith Chapters

Abstract

Hadith chapters are collections of the narrations that quote of what Prophet Muhammad (pbuh) said and preached on Islamic way of living based on the Al Quran. It covers various subjects that concern us as human beings, including wisdom, doctrine, worship and the law especially on the subject of the relationship between Allah and His creatures. In a broader application, hadith chapters are also interpreted with the deeds and acts of Prophet Muhammad (pbuh) and also reports about his companions which the prophet agreed upon. All these three categories are known as the Sunnah. This research investigates the relationships between words in the hadith chapters at the keyword level using a combination technique of text mining and Self Organizing Maps (SOM) cluster analysis to discover frequency of keywords occurred in Hadith chapters and its similarities between different hadith chapters. In this study, we used the hadith documents which were translated into English. The pre-processing steps are necessary in order to eliminate noise and to only keep the useful words. This is an effective and efficient method for Hadith chapters document clustering. The result shows the discovery of the relationships between keywords in the hadith chapters and their relevance. This may give benefits to the Muslims and Islamic scholars to make full use of the Hadith and Sunnah in their daily and also formal practices.