plugin.tt_news ERROR:
No code given

Projekt Data Science Against Climate Change

WiSe 19/20

Modulbereich: Masterprojekt

VAK: 03-MP-902.15


Raum: NN

CP: 30


The project applies advanced visualization and machine learning techniques to expose illegal deforestation that negatively affects climate change.



The farming of beef and soy is one of the main reasons for continuous deforestation, and hence one of the main causes of climate change. With increased meat consumption, there is a growing demand for farmland. Often, this deforestation is illegal. New policy regulations at international level aim to address this issue by promoting deforestation-free supply chains. This means that only agricultural commodities from those farms are being traded within Europe which are not associated with deforestation. In this project, we want to apply data science approaches to analyse and monitor these supply chains and expose
how different actors hide illegal deforestation.

The project will enable students to work on a challenging real-world problem with a large-scale dataset. Based on data from one of Brazil’s biggest federal states, we will explore possibilities for visualising beef-supply chains. We will apply machine learning to detect cattle movement patterns which may be associated with deforestation. Our project is situated in an emerging research area called Computational Social Science (CSS).
Lazer et al. characterised CSS as a field that leverages the capacity to collect and analyse data at scale to examine patterns of individual and group behaviours and to enhance our understanding of individuals and collectives.
A collaboration between social sciences and computer science opens up unprecedented opportunities to understand complex systems. The project will be conducted in collaboration with researchers from Brazil as well as Greenpeace Brazil and Germany. It aims to deliver a system that showcases the potential of data science and data visualization for monitoring purposes to clean supply-chains and exert civil society power.

In the project, participants will learn how to formulate research questions, identify and analyse user requirements, and design and develop visualization and machine learning models. We follow a research-oriented learning paradigm, i.e. the project is self-organised. Participants will not only learn about research in data science but also gain experience in managing a long-term project. Basic programming experience is required to succeed in this project. But the project is mostly about concepts and aimed at anybody that wants to learn more
about data science and machine learning.



Prof. Raoni Rajão (Federal University of Minas Gerais, Brazil)
Greenpeace Germany
Greenpeace Brazil




Prof. Dr. Juliane Jarke

Hendrik Heuer