Text Mining Classification for Translation of Al-Qur'an Verses using the Naive Bayes Classifier

Abstract

The Qur'an is a document and a shield for guidance especially for Muslims (Muslim). In Qur'anic the studying, people have to knowledge at least about the classification in the Qur'an. Which have the function in studying further of a verse and facilitate in searching for verses related to something else or  beetwen verses relationship. One method that can be used for the classification of Qur’an verses is Naive Bayes Classifier method. This is one of the statistical methods that used to perform classification analysis. This paper discusses about formation of document classification models using it. Verse classification is based on the wordlist in the verse. Then class labeling is based on selected words. In this study, the document used is document translation of the verses of Qur'an. The Naive Bayes model is applied to data testing, which is a document of Qur'anic verses that do not yet have a class label. The result of this research are: (1) There are 198 wordlist from result of preproccessing which is used to process formation of Naive Bayes classification model. (2) The prediction results of the new document using 15 models that have been established, there are 47 verses that do not belong to the 15 predetermined categories, while the other 707 verses have been determined in that categories.