Following the recent White House call to action to the tech community to develop techniques that can answer scientific questions related to COVID-19, a team of Vector researchers have released a demo version of their scientific paper search and visualization tool.
Developed by Vector students Duncan Forster and John Giorgi and their advisor Vector Faculty Member Bo Wang, CiteNet indexes the pre-print servers bioRxiv and medRxiv. Pre-prints are completed research papers that have yet to be peer-reviewed, making them the most up-to-date research publically available.
“This tool can make researchers’ lives a lot easier,” says Wang. “Because it is updated daily, it captures the newest, most relevant research.”
CiteNet differs from other academic search tools in that its search is powered by papers, rather than keywords. Using advances in natural language processing, a subset of machine learning, it scans papers for “semantic similarities” and ranks them based on their likely relevance to the papers that were originally inputted.
“The app is still in development,” notes Wang, who says the team has been working on CiteNet for some time. “But we hope to provide a useful, up-to-date literature search tool for members of the scientific community looking to contribute to the global fight against COVID-19.”