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

Abstract

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