The United Nations Framework Convention on Climate Change (UNFCCC) is an international treaty adopted in 1992 at the Earth Summit in Rio. It provides the framework for the climate negotiations between all parties (i.e., countries) that have ratified the Convention. The countries meet every year at the Conference of the Parties (COP), which started in 1995 in Berlin. The 25th COP will be held in December 2019 in Madrid.
Interactions between countries taking part in these negotiations provide a rich environment to study the global competitive dynamics of our world. In this project, we aim at collecting and analyzing a new dataset of collaborations and conflicts between countries, as well as develop models of such dynamics.
This project is part of a collaboration with political scientists at the University of Bern and the University of Zurich.
The participants to every COP (from countries, NGOs, businesses, and official UN agencies) is published online in the form of PDF documents of more than a thousand pages. The first task will be to extract the list of participants, together with some metadata such as their nationality, role, and gender. We will then form a history of all delegations and study how does it evolve over time.
Spicy bit: The oldest documents are scans of type-writer documents, so you will need to apply computer vision techniques to process these.
NGOs do a thorough work of explaining and summarizing negotiations. Their summaries include terms such as "country A opposes country B" and "country C supports country D". The second task will be to extract collaborations and conflicts between countries in the negotiations.
Spicy bit: NGOs also use more complex sentence formulations that require better machine understanding. You will need to use natural language processing techniques to obtain these.
Once the data are obtained, the third task would be to come up with models of the negotiations in order to better understand the dynamics of the process. Can we represent the dynamics as a graph between countries? Can we predict the collaborations and conflicts? These are questions we should be able to develop and answer from a statistical and machine learning viewpoint.
Spicy bit: It is not clear what we will be able the model or not. You will need to come up with new, innovative ideas.
Hard skills
Soft skills
An interest in international politics and/or climate change would help completing a great project :)
Please send your grades and a CV to victor.kristof@epfl.ch.