NoPropaganda at SemEval-2020 Task 11: A Borrowed Approach to Sequence Tagging and Text Classification

TL;DR

This paper presents a sequence tagging and text classification approach for propaganda detection in news articles, utilizing LSTM and transformer models to identify spans and classify propaganda techniques, achieving notable F-scores.

Contribution

It introduces a novel application of autoregressive transformer decoders and span-enveloping techniques for propaganda detection and classification.

Findings

F-score of 44.6% for span detection

Micro-averaged F-score of 58.2% for technique classification

Effective use of borrowed relation extraction methods

Abstract

This paper describes our contribution to SemEval-2020 Task 11: Detection Of Propaganda Techniques In News Articles. We start with simple LSTM baselines and move to an autoregressive transformer decoder to predict long continuous propaganda spans for the first subtask. We also adopt an approach from relation extraction by enveloping spans mentioned above with special tokens for the second subtask of propaganda technique classification. Our models report an F-score of 44.6% and a micro-averaged F-score of 58.2% for those tasks accordingly.