A multi-layer approach to disinformation detection in US and Italian news spreading on Twitter

Abstract

We tackle the problem of classifying news articles pertaining to disinformation vs mainstream news by solely inspecting their diffusion mechanisms on Twitter. This approach is inherently simple compared to existing text-based approaches, as it allows to by-pass the multiple levels of complexity which are found in news content (e.g. grammar, syntax, style). As we employ a multi-layer representation of Twitter diffusion networks where each layer describes one single type of interaction (tweet, retweet, mention, etc.), we quantify the advantage of separating the layers with respect to an aggregated approach and assess the impact of each layer on the classification. Experimental results with two large-scale datasets, corresponding to diffusion cascades of news shared respectively in the United States and Italy, show that a simple Logistic Regression model is able to classify disinformation vs mainstream networks with high accuracy (AUROC up to 94%). We also highlight differences in the sharing patterns of the two news domains which appear to be common in the two countries. We believe that our network-based approach provides useful insights which pave the way to the future development of a system to detect misleading and harmful information spreading on social media.

The COVID-19 Infodemic: Twitter versus Facebook

“…As a benchmark for interpreting the prevalence of lowcredibility content, we also curate a list of 20 more credible information sources. We start from the list provided in a recent Pew Research Center report (Mitchell et al, 2014) and used in a few studies on online disinformation (Pierri et al, 2020a(Pierri et al, , 2020b, and we select popular news outlets that cover the full U.S. political spectrum. These sources have an MBFC factual-reporting level of "Mixed" or higher.…”

The COVID-19 Infodemic: Twitter versus Facebook

“…As a benchmark for interpreting the prevalence of low-credibility content, we also curate a list of 20 high-credibility information sources. We start from the list provided in a recent Pew Research Center report [21] and used in a few studies on online disinformation [31,32], and we select popular news outlets that cover the full U.S. political spectrum. In addition, we include the websites of two organizations that acted as authoritative sources of COVID-19 related information, namely the Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO).…”

Fact checking audiovisual en la era de la posverdad. ¿Qué significa validar una imagen?

“…En paralelo a la más que espinosa lectura textual de la imagen, y de nuevo en términos estrictos de verificación, lo que llama nuestra atención es que la fotografía manipulada pertenece a nuestro tiempo con escalofriante precisión por la ya citada quiebra en la fiabilidad de la fuente -un partido político que se dice democrático, esto es, preocupado por la responsabilidad de su relación con la verdad para el pueblo-, pero no menos importante, su precisa transmisión mediante una red social (Twitter), cuyo compromiso con las mínimas normas de formación de la ciudadanía son, cuanto menos, discutibles (Keller et al, 2020;Pierri, Piccardi and Ceri, 2020) . Como ya señaló Todorov en su diagnóstico sobre los enemigos de la democracia:…”