An AI was tasked with combing through 500,000 studies and white papers to find the most relevant information to help end world hunger. Could machine learning help us bring an end to world hunger? (Image credit: Shuttershock)
It’s hard to believe that world hunger is still a prevalent issue. Research shows up to 60 million people have experienced hunger in the last five years to an estimated 690 million people. United Nations hopes to end world hunger this decade, but first, they have to figure out how, which requires hours upon hours of research. Researchers with Ceres2030, a group of climate, social, and agricultural scientists and economists, are looking for the best ways to solve the problem, and they’re using machine learning to help them get through mountains of research.
To help them get through 500,000 studies and white papers about the world’s food systems, the team relied on artificial intelligence to shorten the process. After being taught what to look for, the off-the-shelf algorithms aided researchers in combing through pieces on agricultural practices and development interventions to improve or reduce hunger. It took a week for the AI to locate and analyze the most useful pieces of research.
With the help of AI along with another analysis by the UN Food and Agricultural Organization and German Center for Development Research, results show that the world needs to dedicate $14 billion per year this decade to end hunger. That’s roughly two percent of what the U.S. annually spends on the military. Machine learning also highlighted a weakness in how research is classified. “Gray literature,” white papers and policy briefs are usually buried on outdated agency websites that lack basic features, like the option to select and download multiple files at once. The information may be out there, but it’s not easily accessible, which needs to change.
The AI analysis also showed where the $14 billion could be placed to get the most out of the aid, such as helping farmers in water-scarce areas invest in livestock and improving access to mobile phone data networks. The latter provides access to the weather forecast to help farmers know when to apply fertilizer between rains to minimize waste, while the former can help improve productivity overall.
Ending world hunger is a huge task. The research the AI was able to locate is essential, but it’s just a small step forward. There’s still a lot of work, research, and studies that need to happen before we come close to solving this issue. But with machine learning doing some of the heavy lifting, it gets us a step closer to reaching this goal.
Have a story tip? Message me at: cabe(at)element14(dot)com