Global Biodefense May 12, 2020
Every day, hundreds of scientific papers about COVID-19 come out, in both traditional journals and non-peer-reviewed preprints. Researchers at Lawrence Berkeley National Lab are using the latest artificial intelligence techniques to build COVIDScholar, a search engine dedicated to COVID-19. It includes tools that pick up subtle clues like similar drugs or research methodologies to recommend relevant research to scientists. AI can capture latent scientific knowledge from text, making connections that humans missed. They built web scrapers that collect new papers as they are published from a wide variety of sources, making them available on the website within 15 minutes of their appearance online. It cleans the data, fixing mistakes in formatting and comparing the same paper from multiple sources to find the best version and tag it with subject categories…read more.