A group of artificial intelligence professionals headed by Stefan Feuerriegel, Director of LMU’s Institute of Artificial Intelligence in Management, is bringing transparency to global development assistance. The researchers created an artificial intelligence system that categorizes assistance initiatives more thoroughly than previously possible and allows for improved project monitoring. The study’s results were published in Nature Sustainability.
“Using our methodology, it is now feasible to track global development assistance initiatives based on a number of criteria, including factors that have never been considered before, such as climate change mitigation. This allows us to spot geographical and temporal disparities as well as any gaps “According to Stefan Feuerriegel. “Our methodology may assist policymakers in making evidence-based choices that are consistent with the United Nations’ Sustainable Development Goals (SDGs).”
AI gives transparency, which is a major milestone
Using artificial intelligence, the LMU team grouped 3.2 million charity programs from 2000 to 2019. These projects have received a combined investment of 2.8 trillion dollars. The projects were assigned to distinct topic groupings using artificial intelligence. “For the first time,” Feuerriegel writes, “this detailed categorisation highlights the critical need for study into greenhouse gas emissions and maternal health care.” Simultaneously, it is now feasible to locate geographic areas where some elements have hitherto been overlooked.
Development assistance encompasses a broad range of activities financed by a variety of organizations. It may take the shape of material contributions, cash donations, training, and even technical assistance, for example. International organizations as well as smaller national sponsors offer funding. “Given the amount of money provided for development assistance, it’s critical to maintain track of where and how it’s spent across the world. This is the only sensible approach to coordinate undertakings on a worldwide scale “According to Feuerriegel. “In the past, the mechanisms used to track projects in this manner were woefully insufficient, prone to time delays, and burdened by high bureaucratic overheads.”
As a result, the LMU researchers are using a machine learning framework to completely gather and evaluate worldwide development assistance efforts. This algorithm was trained using millions of unique project descriptions. The program created a complete and detailed classification of 173 worldwide assistance activity groupings based on these textual descriptions, including education and nutrition, as well as biodiversity. As a result, LMU researchers have achieved a significant breakthrough in improving data-based analytics to promote long-term development.