News2PubMed: A Browser Extension for Linking Health News to Medical Literature

Jun Wang, Bei Yu

Abstract: This demo system presents a browser extension that allows the reader of a health news article to quickly retrieve related medical/health research papers. This system can help news editors and readers fact-check health news for incorrect or exaggerated claims, such as making causal claims from correlational findings or inference of animal studies to humans. Linking health news to the original research papers is not a trivial task, as links are largely missing in science news reports. To link health news to medical literature, our system includes a new named-entity recognition function to extract journal names, and a new Elasticsearch-based search engine to incorporate rich metadata into the search strategy. This paper also introduces a new dataset for evaluating the performance of the proposed search system.