Classification of Indonesian News using LSTM-RNN Method

Abstract

News categorization has the aim of categorizing news into certain categories. In this paper, we build a machine learning model to categorize Indonesian news. One of the best methods for predicting large text sets is the Recurrent Neural Network (RNN) algorithm with Long-Short Term Memory (LSTM) architecture. In previous studies, the use of the LSTM-RNN method has a high level of accuracy for classifying news in English. For further exploration, in this study, a dataset to train and test the Indonesian news application model from the Jakartaresearch and web scraping from Kompas.com is used. Based on the experiment for the LSTM-RNN model, the final score of accuracy was 93%, the recall score was 91.8%, the precision score was 92.4%, and the Fl-Score score was 91.8%s. 17 news predictions from Detik.com have 100% accurate results predicting the correct category.