Automatic detection of influential actors in disinformation networks

Abstract

The weaponization of digital communications and social media to conduct disinformation campaigns at immense scale, speed, and reach presents new challenges to identify and counter hostile influence operations (IOs). This paper presents an end-to-end framework to automate detection of disinformation narratives, networks, and influential actors. The framework integrates natural language processing, machine learning, graph analytics, and a network causal inference approach to quantify the impact of individual actors in spreading IO narratives. We demonstrate its capability on real-world hostile IO campaigns with Twitter datasets collected during the 2017 French presidential elections and known IO accounts disclosed by Twitter over a broad range of IO campaigns (May 2007 to February 2020), over 50,000 accounts, 17 countries, and different account types including both trolls and bots. Our system detects IO accounts with 96% precision, 79% recall, and 96% area-under-the precision-recall (P-R) curve; maps out salient network communities; and discovers high-impact accounts that escape the lens of traditional impact statistics based on activity counts and network centrality. Results are corroborated with independent sources of known IO accounts from US Congressional reports, investigative journalism, and IO datasets provided by Twitter.

Sentinel node approach to monitoring online COVID-19 misinformation

“…Importantly, this analysis addresses the broader case of misinformation, which we consider to be false or misleading content regardless of intent, as opposed to the subset of misinformation known as disinformation which refers to intentionally disseminated false or misleading content within a target group to advance an agenda or to cause harm. The described framework could be used in conjunction with emerging techniques in the detection of influence operations, such as those developed by Smith et al [56], in order to explore the extent to which such actors drive meaningful narrative shifts across the social media ecosystem.…”

Identification and characterization of misinformation superspreaders on social media

“…These results, though encouraging, rely on model-based simulations and decade-old data. More recent work has proposed methods for identifying fake news spreaders and influential actors within disinformation networks that rely on deep neural networks and other machine learning algorithms [26,46]. These methods, however, are complex and hard to interpret.…”

Maximum Entropy Networks Applied on Twitter Disinformation Datasets

“…concerning the 2016 U.S. elections, and despite their availability, the datasets from the Twitter information operations report have not been much studied. One exception is a study that built a semisupervised ensemble-tree classifier model to detect influential actors in a disinformation network [37]. We use these datasets as a ground truth (curated by the social media platform) to detect disinformation operations.…”