Fighting Disinformation Warfare with Artificial Intelligence: Identifying and Combatting Disinformation Attacks in Cloud-based Social Media Platforms

摘要

Following well-documented Russian interference in the 2016 U.S. Presidential election and in the Brexit referendum in the U.K., law enforcement, intelligence agencies and social network providers worldwide have expressed growing interest in identifying and interdicting disinformation warfare. This paper reports on a research project being conducted by the International CyberCrime Research Centre (ICCRC) at Simon Fraser University (Canada) in cooperation with the Department of Information and Computer Sciences at the University of Strathclyde (Scotland). The research project involves the development of a method for identifying hostile disinformation activities in the Cloud. Employing the ICCRC's Dark Crawler, Strathclyde's Posit Toolkit, and TensorFlow, we collected and analyzed nearly three million social media posts, examining "fake news" by Russia's Internet Research Agency, and comparing them to "real news" posts, in order to develop an automated means of classification. We were able to classify the posts as "real news" or "fake news" with an accuracy of 90.12% and 89.5%, using Posit and TensorFlow respectively.