People Still Care About Facts: Twitter Users Engage More with Factual Discourse than Misinformation--A Comparison Between COVID and General Narratives on Twitter
arXiv:2012.02164
Abstract
Misinformation entails the dissemination of falsehoods that leads to the slow fracturing of society via decreased trust in democratic processes, institutions, and science. The public has grown aware of the role of social media as a superspreader of untrustworthy information, where even pandemics have not been immune. In this paper, we focus on COVID-19 misinformation and examine a subset of 2.1M tweets to understand misinformation as a function of engagement, tweet content (COVID-19- vs. non-COVID-19-related), and veracity (misleading or factual). Using correlation analysis, we show the most relevant feature subsets among over 126 features that most heavily correlate with misinformation or facts. We found that (i) factual tweets, regardless of whether COVID-related, were more engaging than misinformation tweets; and (ii) features that most heavily correlated with engagement varied depending on the veracity and content of the tweet.
arXiv comment22 pages
References in corpus (12)
- Scikit-learn: Machine Learning in Python
- Sentiment of Emojis
- Scalable and Generalizable Social Bot Detection through Data Selection
- A first look at COVID-19 information and misinformation sharing on Twitter
- CoAID: COVID-19 Healthcare Misinformation Dataset
- COVID-19 on Social Media: Analyzing Misinformation in Twitter Conversations
- Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
- TweetsCOV19 -- A Knowledge Base of Semantically Annotated Tweets about the COVID-19 Pandemic
- Disinformation and Misinformation on Twitter during the Novel Coronavirus Outbreak
- Characterizing COVID-19 Misinformation Communities Using a Novel Twitter Dataset
- Ginger Cannot Cure Cancer: Battling Fake Health News with a Comprehensive Data Repository
- Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks