Knowing Opposing Arguments in Persuasive Essays Using Random Forest Classifier

Argumentation mining is a relatively new field of research in the perspective of computational linguistics. It can be used to improve the quality of arguments in persuasive essays by detecting whether in a persuasive essay there are opposing arguments or not. This is because the importance of opposing arguments in persuasive essays that can improve the quality of arguments, precision and author's claims. This research adapted the research conducted by Stab and Gurevych who used the SVM classification method with an accuracy 75,6% and macro F1-score 0,734. While in this research used the Random Forest method and get accuracy of 85,125% and macro F1-score of 0,841 by using three features, namely unigram, production rules, and adversative transitions.