The problem is that after machine reading, computers have difficulties figuring out even simple statements, so the scientists who devised the software program has used probability to decipher the text.
Christopher Ré, a professor of computer sciences, project managed the software development in a bid to quickly summarise, collate and index the mountain of data produced by scientists worldwide.
They tested their system called PaleoDeepDive against human scientists manually entering data into the Paleobiology database. The software mimics the action of the human scientists and the machine accessed tens of thousands of articles.
Shanan Peters, a professor of science at UWM, said: “Ultimately, we hope to have the ability to create a computer system that can do almost immediately what many geologists and paleontologists try to do on a smaller scale over a lifetime: read a bunch of papers, arrange a bunch of facts, and relate them to one another in order to address big questions”.