Accelerating evidence-informed decision-making for the Sustainable Development Goals using machine learning

摘要: The United Nations Sustainable Development Goal 2 (SDG 2) is to achieve zero hunger by 2030. We have designed Persephone, a machine learning model, support diverse volunteer network of 77 researchers from 23 countries engaged in creating interdisciplinary evidence syntheses SDG 2. Such syntheses, whatever the specific topic, assess original studies determine effectiveness interventions. By gathering and summarizing current providing objective recommendations they can be valuable aids decision-makers. However, are time-consuming; estimates range 18 months three years produce single review. Persephone analysed 500,000 unstructured text summaries prominent sources agricultural research, determining with 90% accuracy subset that would eventually selected expert researchers. demonstrate models invaluable placing into hands policymakers. Evidence produced scientific literature important tools for Producing such highly time- labour-consuming but help as already demonstrated health medical sciences. This Perspective describes learning-based framework specifically area tackling UN 2: