A machine learning system trained on scholarly journals could correct Wikipedia's gendered under-representation problem

Yeah, I’d love to see how they avoid false hits on that.