Team Fernando-Pessa at SemEval-2019 Task 4: back to basics in Hyperpartisan News Detection

This paper describes our submission1 to theSemEval 2019 Hyperpartisan News Detectiontask. Our system aims for a linguistics-baseddocument classification from a minimal setof interpretable features, while maintaininggood performance. To this goal, we followa feature-based approach and perform severalexperiments with different machine learningclassifiers. On the main task, our modelachieved an accuracy of 71.7%, which wasimproved after the task's end to 72.9%. Wealso participate in the meta-learning sub-task,for classifying documents with the binary classificationsof all submitted systems as input,achieving an accuracy of 89.9%