With Google as an indispensable tool omnipresent in everyday life, it’s easy to feel like we have all the information that there is available at our fingertips. Yet that’s not the case in the scientific world, which seems to be suffering from a bad case of information overload.

It’s estimated that of the millions of academic papers published each year, only around half are read by an audience bigger than the author and the editors themselves. Vast pools of information that could hold the key to important developments are concealed amidst the deluge.

“What if a cure for an intractable cancer is hidden within the tedious reports on thousands of clinical studies?” says Oren Etzioni, CEO of the Seattle-based Allen Institute for Artificial Intelligence (AI2). Their recently launched search engine, Semantic Scholar, trudges through the multitude of information so others don’t have to.

Thanks to software that is able to process and understand natural language, the service can look deep inside articles, extracting nuanced concepts where other engines would have only matched keywords. The most important papers for the user’s query are identified, as are the connections between them and their key phrases.

Additional sense is made by establishing which references in a paper are truly influential, as opposed to those included solely for background or comparison. By applying filters, searches can be narrowed down further according to the conference at which the paper was presented, or the journal in which it was published. For the time being Semantic Scholar is able to digest around three million papers on computer science with its reach soon to be broadened to other academic areas including neuroscience.

Laura Humphries is a Barcelona/London-based writer interested in matters of urbanism, international development, human rights and more.