Characterisation and analysis of the international agro-food trade network

---

Authors

• Arganaraz Juan Carlos (University of Padua)
• Capettini Hilario (University of Padua)
• Carmona Gerardo (University of Padua)
• Rinaldi Luca (University of Padua)

Abstract

Diverse systems that may seem too elaborate at first, can be modelled in an elegant and efficient fashion by means of graphs. Once the theoretical and empirical perspectives on such models are combined, the powerful concepts and techniques of network science emerge. The study of international trade profits extensively from such an approach. Global food safety and sustainability is a growing concern; understanding the underlying structures at different scales is thus essential to promote well-being. In this work, we model international agro-food trade as a weighted, directed graph, aiming to extract insightful results from the application of network science methods. At first we perform a full characterisation of the network by the means of the general proprieties and different metrics. Then we deeply investigate the eventual emerging communities and their evolution over time, first with a single-layer approach and then with a multi-layer ones. Finally we analyse the Robustness of the network exploiting different metrics.

• The complete report of this project can be found here

• The presentation of this project can be found here

Implementations

---

Data mining & Processing	Centrality & other Metrics

Single-Layer Community Detection	Multi-Layer Community Detection	Robustness Analysis