State media warning labels can counteract the effects of foreign misinformation

Platforms are increasingly using transparency, whether it be in the form of political advertising disclosures or a record of page name changes, to combat disinformation campaigns. In the case of state-controlled media outlets on YouTube, Facebook, and Twitter this has taken the form of labeling their connection to a state. We show that these labels have the ability to mitigate the effects of viewing election misinformation from the Russian media channel RT. However, this is only the case when the platform prominently places the label so as not to be missed by users.

A Multi-Modal Method for Satire Detection using Textual and Visual Cues

Satire is a form of humorous critique, but it is sometimes misinterpreted by readers as legitimate news, which can lead to harmful consequences. We observe that the images used in satirical news articles often contain absurd or ridiculous content and that image manipulation is used to create fictional scenarios. While previous work have studied text-based methods, in this work we propose a multi-modal approach based on state-of-the-art visiolinguistic model ViLBERT. To this end, researchers created a new dataset consisting of images and headlines of regular and satirical news for the task of satire detection. The researchers also fine-tune ViLBERT on the dataset and train a convolutional neural network that uses an image forensics technique. Evaluation on the dataset shows that our proposed multi-modal approach outperforms image-only, text-only, and simple fusion baselines.

Validating Social Media Monitoring: Statistical Pitfalls and Opportunities from Public Opinion

Social media are a promising new data source for real-world behavioral monitoring. Despite clear advantages, analyses of social media data face some challenges. In this paper, we seek to elucidate some of these challenges and draw relevant lessons from more traditional survey techniques. Beyond standard machine learning approaches, we make the case that studies that conduct statistical analyses of social media data should carefully consider elements of study design, providing behavioral examples throughout. Specifically, we focus on issues surrounding the validity of statistical conclusions that may be drawn from social media data. We discuss common pitfalls and techniques to avoid these pitfalls, so researchers may mitigate potential problems of design.