Shanan Peters, University of Wisconsin, firstname.lastname@example.org
Synthesizing published data and information is critical to addressing a wide range of questions in the Earth and Life sciences. Advances in machine reading and learning offers the potential to automate and improve key steps in the process of discovering and aggregating published information, but considerable computing and data infrastructure is required to power such a system. Here I describe xDD (formerly GeoDeepDive), our effort to provide such a computing and data resource. Our recent work on AI-based approaches to text, table, figure, and equation recognition and extraction will be presented.
Note: presentation was in Keynote so we do not have the slides.