A Search Engine for Discovery of Scientific Challenges and Directions

Keeping track of scientific challenges, advances and emerging directions is a fundamental part research. However, researchers face flood papers that hinders discovery important knowledge. In biomedicine, this directly impacts human lives. To address problem, we present novel task extraction search challenges directions, to facilitate rapid knowledge discovery. We construct release an expert-annotated corpus texts sampled from full-length papers, labeled with semantic categories generalize across many types directions. focus on large interdisciplinary work relating the COVID-19 pandemic, ranging biomedicine areas such as AI economics. apply model trained our data identify build dedicated engine. experiments 19 clinicians using system, outperform popular engine in assisting Finally, show models resource wider biomedical domain highlighting its broad utility. make data, publicly available.