Artificial neural networks could power up curation of natural history collections
02 / 11 / 2017, Eureka Alert
Fed with new knowledge for centuries, natural history collections contain critical data for many scientific endeavors. While recent efforts in mass digitization have already provided unprecedented insight by generating large datasets from these collections, a new pilot project -- one of the first of its kind -- suggests that the key to efficiently studying these data might lie in the new-age deep learning techniques. The research article is published in the open-access Biodiversity Data Journal.