Misinformation Adoption or Rejection in the Era of COVID-19

While billions of COVID-19 vaccines have been administered, too many people remain hesitant. Twitter, with its substantial reach and daily exposure, is an excellent resource for examining how people frame their vaccine hesitancy and to uncover vaccine hesitancy profiles. In this paper we expose our processing journey from identifying Vaccine Hesitancy Framings in a collection of 9,133,471 original tweets discussing the COVID-19 vaccines, establishing their ontological commitments, annotating the Moral Foundations they imply to the automatic recognition of the stance of the tweet authors toward any of the CoVaxFrames that we have identified. When we found that 805,336 Twitter users had a stance towards some CoVaxFrames in either the 9,133,471 original tweets or their 17,346,664 retweets, we were able to derive nine different Vaccine Hesitancy Profiles of these users and to interpret these profiles based on the ontological commitments of the frames they evoked in their tweets and on value of their stance towards the evoked frames.

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

From Hesitancy Framings to Vaccine Hesitancy Profiles: A Journey of Stance, Ontological Commitments and Moral Foundations

Weinzierl

Harabagiu

2022

“…One class involves computational methods to detect misinformation . These computerbased approaches have shown remarkable success, leveraging signals such as sharing patterns (Rosenfeld, Szanto, and Parkes 2020), text features (Granik and Mesyura 2017), account activity (Breuer, Eilat, and Weinsberg 2020), user stance (Weinzierl, Hopfer, and Harabagiu 2021) and vi-sual features for website screenshots (Abdali et al 2021). However, the nuanced nature of truth, the limited availability of labeled training data (Rubin, Chen, and Conroy 2015;Bozarth, Saraf, and Budak 2020), and the nonstationarity problem whereby the signatures of misinformation can change rapidly (e.g.…”

Section: Approaches To Detecting Misinformationmentioning

confidence: 99%

“…In addition, methods employing professional fact-checkers suffer from a general lack of bi-partisan trust: one study from the Pew Research Center found that 70% of Republicans and 48% of Americans say fact-checking efforts tend to "favor one side" (Walker and Gottfried 2019). Another proposed alternative is fully algorithmic methods for detecting misinformation Abdali et al 2021;Weinzierl, Hopfer, and Harabagiu 2021) (see Related Work for a review). However, these methods fundamentally struggle to keep up with the the non-stationarity of misinformation, and require ground-truth labeling.…”

Section: Introductionmentioning

confidence: 99%

Do Explanations Increase the Effectiveness of AI-Crowd Generated Fake News Warnings?

Epstein

Foppiani

Hilgard

et al. 2022

Social media platforms are increasingly deploying complex interventions to help users detect false news. Labeling false news using techniques that combine crowd-sourcing with artificial intelligence (AI) offers a promising way to inform users about potentially low-quality information without censoring content, but also can be hard for users to understand. In this study, we examine how users respond in their sharing intentions to information they are provided about a hypothetical human-AI hybrid system. We ask i) if these warnings increase discernment in social media sharing intentions and ii) if explaining how the labeling system works can boost the effectiveness of the warnings. To do so, we conduct a study (N=1473 Americans) in which participants indicated their likelihood of sharing content. Participants were randomly assigned to a control, a treatment where false content was labeled, or a treatment where the warning labels came with an explanation of how they were generated. We find clear evidence that both treatments increase sharing discernment, and directional evidence that explanations increase the warnings' effectiveness. Interestingly, we do not find that the explanations increase self-reported trust in the warning labels, although we do find some evidence that participants found the warnings with the explanations to be more informative. Together, these results have important implications for designing and deploying transparent misinformation warning labels, and AI-mediated systems more broadly.

“…There have been different approaches proposed to deal with this growing challenge. For example, researchers have been developing effective methods to automatically detect online misinformation (Gatto, Basak, and Preum 2023;Weinzierl, Hopfer, and Harabagiu 2021;Shu, Wang, and Liu 2019). This is used to support expert fact-checkers who would be otherwise overwhelmed by the amount of content to be fact-checked.…”

Section: Introductionmentioning

confidence: 99%

On the Role of Large Language Models in Crowdsourcing Misinformation Assessment

Xu,

Han,

Sadiq

et al. 2024

The proliferation of online misinformation significantly undermines the credibility of web content. Recently, crowd workers have been successfully employed to assess misinformation to address the limited scalability of professional fact-checkers. An alternative approach to crowdsourcing is the use of large language models (LLMs). These models are however also not perfect. In this paper, we investigate the scenario of crowd workers working in collaboration with LLMs to assess misinformation. We perform a study where we ask crowd workers to judge the truthfulness of statements under different conditions: with and without LLMs labels and explanations. Our results show that crowd workers tend to overestimate truthfulness when exposed to LLM-generated information. Crowd workers are misled by wrong LLM labels, but, on the other hand, their self-reported confidence is lower when they make mistakes due to relying on the LLM. We also observe diverse behaviors among crowd workers when the LLM is presented, indicating that leveraging LLMs can be considered a distinct working strategy.