Towards Detecting Cascades of Biased Medical Claims on Twitter
This addresses the issue of misinformation in public health for social media users and researchers, though it is incremental as it applies existing models to a new domain.
The study tackled the problem of identifying and measuring the spread of biased medical claims on Twitter, finding that such claims circulate faster and further than unbiased ones.
Social media may disseminate medical claims that highlight misleading correlations between social identifiers and diseases due to not accounting for structural determinants of health. Our research aims to identify biased medical claims on Twitter and measure their spread. We propose a machine learning framework that uses two models in tandem: RoBERTa to detect medical claims and DistilBERT to classify bias. After identifying original biased medical claims, we conducted a retweet cascade analysis, computing their individual reach and rate of spread. Tweets containing biased claims were found to circulate faster and further than unbiased claims.