2017/01/16 by Mauro Conti, Daniele Lain, Conti, Mauro +7 · 2 citations
Social Sciences · Computer Science · #Misinformation and Its Impacts #Spam and Phishing Detection #Media Influence and Politics
paper · pdf · doi:10.48550/arxiv.1701.04221
Given the huge impact that Online Social Networks (OSN) had in the way people\nget informed and form their opinion, they became an attractive playground for\nmalicious entities that want to spread misinformation, and leverage their\neffect. In fact, misinformation easily spreads on OSN and is a huge threat for\nmodern society, possibly influencing also the outcome of elections, or even\nputting people's life at risk (e.g., spreading "anti-vaccines" misinformation).\nTherefore, it is of paramount importance for our society to have some sort of\n"validation" on information spreading through OSN. The need for a wide-scale\nvalidation would greatly benefit from automatic tools.\n In this paper, we show that it is difficult to carry out an automatic\nclassification of misinformation considering only structural properties of\ncontent propagation cascades. We focus on structural properties, because they\nwould be inherently difficult to be manipulated, with the the aim of\ncircumventing classification systems. To support our claim, we carry out an\nextensive evaluation on Facebook posts belonging to conspiracy theories (as\nrepresentative of misinformation), and scientific news (representative of\nfact-checked content). Our findings show that conspiracy content actually\nreverberates in a way which is hard to distinguish from the one scientific\ncontent does: for the classification mechanisms we investigated, classification\nF1-score never exceeds 0.65 during content propagation stages, and is still\nless than 0.7 even after propagation is complete.\n