<![CDATA[The Al Qur'an is a holy book of Muslims which is a guide to life for all mankind. Studying and understanding the translation of the Al-Quran is not easy, one way that can be done is to classify the translation of Al-Quran verses into existing topics. This research uses Naïve Bayes and LSTM methods in the classification process. The data used comes from translation data of the Al-Quran in Indonesian which has been labeled based on multi-class classification. One of the main problems faced is data imbalance. To overcome this problem, data balancing, text preprocessing, feature construction and feature extraction processes were carried out using the Bag of Words (BoW) and TF.IDF techniques. The research results indicate that the most optimal Naïve Bayes model achieved an average accuracy of 55.39% on test data from juz 30, 61.59% on test data from juz 10-20, and 59.53% on test data from juz 25-28. Meanwhile, the most optimal LSTM model yielded an accuracy of 58.02% on test data from juz 30, 59.64% on test data from juz 10-20, and 58.59% on test data from juz 25-28. The main aim of this research is to improve classification performance and compare the accuracy between naïve Bayes and lstm.]]>
Copyrights © 2024
