Scientists and tech-leaders from five of the leading American organisations such as National Institute of Health’s National Library of Medicine (NIHNLM), Microsoft, and Chan Zuckerberg Initiative have collaborated to release an open dataset of nearly 30,000 scientific articles hoping that artificial intelligence (AI) experts might be able to use the data to help combat the spread of COVID-19 infections. The dataset will be made machine-readable so that AI can easily grasp the information. 

The CORD-19 dataset, short for COVID-19 Open Research Data set, was released after the Trump administration urged the tech community to engage AI to combat the pandemic. “Decisive action from America’s science and technology enterprise is critical to prevent, detect, treat, and develop solutions to COVID-19,” said Michael Kratsios, U.S. Chief Technology Officer, at a White House press meeting. 

While the dataset is reported to be the most extensive of its’ kind, currently only 13,000 articles are machine-readable. The dataset was released in the hopes of spurring America’s AI experts to “develop new techniques for mining data and text that could help answer some of the most pressing questions about the novel coronavirus and the disease it causes,” as reported by Stat News. 

The five American organisations will be collaborating to leverage their strengths to make CORD-19 completely machine-readable and accessible to the broader scientific community. The NIHMLM has provided the literature content that was transformed into the machine-readable format by Allen Institute for AI and curated by Microsoft’s literature curation tools. The Chan Zuckerberg Initiative provided access to articles that have been posted on preprint servers but not yet peer-reviewed. Georgetown University’s Centre for Security and Emerging Technology coordinated the initiative. The dataset was created on the request of the White House’s Office of Science and Technology Policy.