AI and Climate Information Needs in Africa

Afra Amini, Christian Buck, HANNAH BROWN, Jannis Bulian, Michelle Chen Huebscher, Massimiliano Ciaramita, Sonali Das, Ben Gaiarin, Chris GORDON, Rangan Gupta, Kenny Kutu, Doreen Larkailey Lartey, Markus Leippold, Leslie Leuenberger, Mark Atta Mensah

Keywords: ai, africa, climate, climate change, llm

Abstract:

What are Africa's climate information needs? How do they relate to global concerns? How can AI help? This paper begins exploring these topics, with a specific focus on Ghana and South Africa.

Our study involves climate-related questions from Google Trends and identifies macro clusters related to the broad themes of understanding climate change, its causes, and potential solutions. A finer-grained analysis reveal country-specific topical profiles, including pronounced concerns, in the two African countries, about the disruptive consequences on land and economy. We also begin to probe the ability of large language models to answer some of these questions. Preliminary experiments provide somewhat contradictory evidence, indicating that this is a subtle and complex issue that deserves more research.

Submission Category: Machine learning algorithms

Submission Number: 62