Rumor Stance Classification: A Case Study on the Propagation of Political Rumors on the Algerian Online Social Space

Abstract

Social media is extensively an important means for spreading and sharing news. Nevertheless, the lack of mechanisms for verifying posted information has given rise to a widespread of fake news and rumors. This paper particularly examined a real case of political rumor spreading that marked the Algerian online space for a few months between the end of the year 2020 and the beginning of 2021. It concerns the rumors that have circulated on the social platform YouTube about the health of the President of the Algerian Republic Abdelmadjid Tebboune. By this contribution, we aimed to detect the stances of the Algerian online public towards these rumors by following a Transfer Learning approach. The experiments on the collected data reported good performance of monolingual transformers compared to multilingual transformers in the task of rumor stance classification.