With the ongoing coronavirus pandemic, the amount of scientific literature regarding the virus has exploded. Tens of thousands of scientific papers have been published since the inception of the virus. Although this may seem like a promising direction, the sheer number of papers being produced can actually be quite problematic. As a result of this, many researchers have joined together to find ways to maximize coronavirus research by creating SemViz, an artificial intelligence algorithm that is capable of sorting through large numbers of scientific papers.
The amount of scientific literature published is a hopeful direction to finding a cure for the coronavirus. Despite how favorable this might seem, it is ironically making it more difficult to find patterns in research. With current technology, researchers would have to spend countless hours reading through individual papers to get any meaningful results, and this is more problematic since these scientific papers analyze the coronavirus in various countries. Manually searching through papers is time-consuming and difficult, and as a result, can make it very hard to find a cure.
James Pustejovsky, a researcher at Brandeis, along with researchers at Harvard, Tufts, University of Illinois, and Vassar College, worked to create what is known as SemViz, which uses semantic visualization, a technique to gather meaning within data, to help find correlations in coronavirus literature. By doing so, biologists and other researchers can find common trends present in a vast array of papers, in this case, finding patterns within certain proteins and genes. The group of researchers, in addition to the semantic visualization, created a convenient search engine feature.
Enabling researchers to sort through some 50,000 papers and helping biologists to quickly study regulators, inhibitors, and genes connected to the virus, this new scientific tool might significantly speed the race to finding a cure for the current coronavirus pandemic.