Moving forward by looking backward

UMSI assistant professor Eytan Adar has been selected to join a large team of academics and scientists from across the U.S. who are working to automatically detect emerging scientific and technical capabilities by analyzing vast collections of technical, patent, and scientific literature.

The team’s goal is to discover whether past patterns can point the way to future discoveries. This research is funded through an award from the Intelligence Advanced Projects Research Activity (IARPA) under its Foresight and Understanding from Scientific Exposition (FUSE) Program.

Start date: 5/17/2012
End date: 5/16/2016

Read More

Because of the overwhelming quantity of scientific and technical information now being published around the world, it is virtually impossible, even for domain experts, to stay current, let alone to see patterns and connections. A research goal of the program is to uncover and understand new scientific and technical breakthroughs by utilizing algorithms and digital analysis to identify emerging technology. 

Adar and his graduate students will be part of SRI’s “Copernicus” team on the IARPA FUSE Program. Copernicus is a five-year project managed by SRI International, an independent nonprofit research institute. The Copernicus system will automatically analyze the text of hundreds of millions of pages of scientific, technical, and patent documents in a variety of languages published in the past three decades to see if there is a way to detect emerging technology long before it rises to public consciousness. The scientists will test their model’s ability to detect the emergence of technologies that have since come into existence. If successful, future programs could identify topics that seem to be emerging in the literature and then “nominate” which findings are worth handing off to human analysts for further study.


Network Analysis of Emerging Scientific and Technological Innovations, IARPA Fuse Program: $824,000


IARPA’s Fuse Program seeks to develop automated methods that aid in the systematic, continuous, and comprehensive assessment of technical emergence using information found in published scientific, technical, and patent literature.