Hierarchical multilabel classification for Indonesian news articles

Abstract

Multilabel news classification is needed since numerous news could be labeled with two or more categories. This research is conducted in the account of Indonesian news, because researches in this area are still rare. The classification method used in this paper is called hierarchical multilabel classification and is used to decide the class and subclass of every single news using problem transformation approach. The aim of this study is to discover whether hierarchical multilabel classification is suitable to be implemented Indonesian news. Based on the experiment conducted, Naïve Bayes algorithm gives the best performance when it applied to construct the model.