newsSweeper at SemEval-2020 Task 11: Context-Aware Rich Feature Representations For Propaganda Classification

This paper describes our submissions to SemEval 2020 Task 11: Detection of\nPropaganda Techniques in News Articles for each of the two subtasks of Span\nIdentification and Technique Classification. We make use of pre-trained BERT\nlanguage model enhanced with tagging techniques developed for the task of Named\nEntity Recognition (NER), to develop a system for identifying propaganda spans\nin the text. For the second subtask, we incorporate contextual features in a\npre-trained RoBERTa model for the classification of propaganda techniques. We\nwere ranked 5th in the propaganda technique classification subtask.\n