Working Papers

This paper proposes the use of synthetic training data generated by large language models
to improve machine learning SDG classifiers. It shows that supplementing existing training data with
synthetic data produced by the ChatGPT tool improves the performance of the SDGClassy classifier.
This addition of synthetic data is especially useful in building SDG classifiers given the limited availability
of properly labeled data and the complex, interconnected nature of the SDGs. Synthetic data thus enables
more effective machine-learning applications in this context.

Implementation of the Agenda 21 bifurcated into two tracks. While the economic and social development agenda gelled into the Millennium Development Goals (MDGs), the environmental protection agenda moved along a different track, represented by the United Nations Framework Convention on Climate Change (UNFCCC), Convention on Biodiversity (CBD), etc. This bifurcation also led to very different “domain configurations.” While there were some advantages of this bifurcation, it led to a conflict between the human development and the environmental goals. This paper presents a framework for bringing environment and development together in the UN post-2015 agenda.