AbstractThe study of international relations is becoming more and more relevant in society today. Effective communication between countries is essential for making beneficial relationships and ensuring a safer world as a result. The main goal of this work is not the study and understanding of historical, economic, or political reasons why the relations between the countries are in a certain way but the relations themselves and use them as an input for predictive models.
We consider the entire world as constructed by one large network in which the nodes are the countries and the edges can represent any kind of relationship that the countries can establish between them.
In the elaboration of this thesis, we address two relevant problems, all about international relations. In both cases, we consider the world as a large network composed of countries and different types of relationships between them.
The first work we have created a model that allows us to predict which countries will lead the export market in the coming years in terms of how much they are capable of displacing others from the international market. And in the second work, we introduce a model to predict conflicts between countries using a multilayer network system and machine learning, the main goal of this work is to demonstrate that international conflicts follow social balance.
As we can see, the common element of all these studies has been to analyze the sensitivity of relations between countries, which translated into the language of networks would be the analysis of the change in the structure of a network in the face of the appearance and disappearance of positive or negative links.
|Date of Award||2020|
|Supervisor||Renaud Lambiotte (Supervisor), Michele COSCIA (Co-Supervisor), Joseph WINKIN (President), Timoteo Carletti (Jury) & Jean-Charles DELVENNE (Jury)|
- Complex systems
- multilayer networks
- link prediction