Identifikasi Berita Hoax Berbahasa Indonesia dengan Menggunakan Algoritma C5.0

Abstract

Hoax or deception is misinformation and is dangerous because it confuses human perception by disseminating false information as truth. Dissemination of hoax information usually has various purposes, with the aim of manipulating public opinion. According to data from the Ministry of Communications and Information Technology, approximately 800,000 websites in Indonesia have been attacked for spreading hoaxes or false information. So we need a system that can identify hoax news in Indonesian. The system built applies the C5.0 algorithm, an algorithm that has become an industry standard for generating decision trees because it can handle most types of problems immediately. The data can be divided into two parts, namely training data and test data, 600 training data taken from the Mendeley Data website with the theme "Indonesia Hoax News Detection Dataset, while 250 test data are taken by crawling techniques on online news portals. System testing is done using 250 testing data, the test method uses a confusion matrix. The results of the test are 86.8% accuracy, 13.2% error, 85.8% recall, 93.3% precision and 89.4% f- measure.