<![CDATA[Hoax newsthat contain incorrect (false) information often become public consumption on social media today. This hoax phenomenon raises doubts about information and makes confusion in the community. In this study, experiments conducted aimed at selecting the best algorithm in classifying hoax and non-hoax news with the number of data in 251 articles in Indonesian language (100 hoax articles and 151 non-hoax articles) using text mining method and machine learning based approaches. This research undergoes the text preprocessing phase which consists of tokenizing, case folding, filtering, stopwords removing, stemming and TF-IDF weighting using unigram and bigram combine features before processing it into classification text. The results of this research is the Random Forest algorithm that gets the best accuracy in classifying hoax and non-hoax news compared to the Multilayer Perceptron algorithm, Naïve Bayes, Support Vector Machine, and Decision Tree with an accuracy value of 76.47%.]]>
Copyrights © 2019
