Paul Allen’s non-profit Allen Institute for Artificial Intelligence (AI2) is expanding its flagship product Semantic Scholar to the world of neuroscience.
The smart search engine, which helps scientists hone their queries using cutting-edge A.I. technology, was previously limited to computer science literature. Today, AI2 is launching a biomedical research portal, focused initially on brain science.
Semantic Scholar is designed to help scientists find the right paper faster and more efficiently, said AI2 CEO Oren Etzioni. “If we can do that, we’ve lived up to our promise of A.I. for the common good.”
The platform uses artificial intelligence to put search queries into context, rather than simply scanning its database for keywords. It’s able to predict the intent behind the search. Semantic Scholar also has specialized filters to help scientists further hone their results, including the ability to filter by region of the brain.
Semantic Scholar uses data mining, natural language processing and computer vision to identify and present key elements from research papers on a summary page. For example, when a researcher clicks an article link, Semantic Scholar shows one page with an abstract, related key phrases, figures and tables from the article, references, citations, stats and related publications.
“If you’re dealing with information overload, you want these things to help you cut through the clutter, slice and dice the results,” said Etzioni.
AI2 built the neuroscience product in close collaboration with the Allen Institute for Brain Science, another research organization started by Allen, the Microsoft co-founder.
Currently, most scientists rely on Google Scholar for research, but Etzioni believes Semantic Scholar offers a more sophisticated experience.
“Our goal is to raise the bar,” he said when the first Semantic Scholar product launched. “It would be foolhardy to take on Google … We’re very interested in helping scientists, not because that’s a good way to make money — it ain’t. Because it’s a good way to make the world a better place for all of us.”
Anyone can use the free search engine, which Etzioni and his team of 70 see as a public service of sorts. Since launching Semantic Scholar for computer science literature, 2.5 million people have used it, according to Etzioni.
The institute’s goal is to expand Semantic Scholar to the full library of biomedical research by the end of 2017.
“We’re executing on Paul’s vision to use A.I. to help scientists,” Etzioni said.