Predicting the Factuality of Reporting of News Media Using Observations about User Attention in Their YouTube Channels

Proceedings of Recent Advances in Natural Language Processing, pages 182–189 Sep 1–3, 2021. https://doi.org/10.26615/978-954-452-072-4_022 182 Predicting the Factuality of Reporting of News Media Using Observations About User Attention in Their YouTube Channels Krasimira Bozhanova FMI, Sofia University “St. Kliment Ohridski” Sofia, Bulgaria krasimira.bozhanova@gmail.com Yoan Dinkov FMI, Sofia University “St. Kliment Ohridski” Sofia, Bulgaria yoandinkov@gmail.com Ivan Koychev FMI and GATE, Sofia University “St. Kliment Ohridski” Sofia, Bulgaria koychev@fmi.uni-sofia.bg Maria Castaldo Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, GIPSA-lab, F-38000, France maria.castaldo@grenoble-inp.fr Tommaso Venturini Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, GIPSA-lab, F-38000, France tommaso.venturini@cnrs.fr Preslav Nakov Qatar Computing Research Institute HBKU Doha, Qatar pnakov@hbku.edu.qa Abstract We propose a novel framework for predict- ing the factuality of reporting of news me- dia outlets by studying the user attention cy- cles in their YouTube channels. In particu- lar, we design a rich set of features derived from the temporal evolution of the number of views, likes, dislikes, and comments for a video, which we then aggregate to the channel level. We develop and release a dataset for the task, containing observations of user attention on YouTube channels for 489 news media. Our experiments demonstrate both complementar- ity and sizable improvements over state-of-the- art textual representations. 1 Introduction Disinformation in the news and in social media is perceived as having a major impact on society, e.g., during the 2016 US Presidential election (Grin- berg et al., 2019) and the Brexit referendum (Gor- rell et al., 2018). During the COVID-19 pandemic outbreak, the moral panic (McLuhan, 1964) around online disinformation grew to a whole new level as the first global infodemic.1 Nowadays, fighting disinformation online has been recognized as one of the most important issues societies around the world are facing today. In this paper, we highlight an aspect of disin- formation that is often neglected. Rather than ex- amining the truth-value of individual piece of in- formation, we investigate the general quality of the attention regimes in different news outlets, by analyzing news media’s YouTube channels. 1MIT Technology Review: tinyurl.com/y8oschng YouTube is the largest and the most popular plat- form for video sharing with over two billion users and it is also the second most widely used news source in USA, after Facebook.2 While the plat- form has been scrutinized for the way in which it may amplify marginal and sometimes radical contents (Munn, 2019; Ribeiro et al., 2020), the connection between attention dynamics and disin- formation levels is still largely unexplored. Here, we do not focus on specific videos, but rather on entire YouTube channels of news media and their attention dynamics. In particular, we are interested in differentiating the “attention cycles” (Downs, 1972; Leskovec et al., 2009) of YouTube channels, that is to assess the rapidity and the steep- ness with which their videos rise and fall in the consideration of their audiences. While some out- lets encourage extensive and diverse discussions, other tend to concentrate everyone’s attention on the latest hot-button, thus distracting the public opinion instead of nourishing it (Venturini, …

Content truncated to 500 words