Aplicação de métodos inferenciais a redes bipartidas no estudo do fluxo turístico Europa - Brasil

Abstract

The present Undergraduate Thesis explores an in-depth study of inferential methods applied to bipartite network data, with a practical application to a dataset that records the arrival of tourists from Europe to the twenty-six Brazilian states, as well as the Federal District, exclusively by air travel. This study focuses on understanding the dynamics of European tourist flows to Brazil by analyzing the arrival of tourists over the semesters between the years 2019 and 2022. To achieve this, inferential methods and computational tools will be employed to provide a detailed analysis of the factors influencing this movement, identifying patterns and trends over time. The modeling will be conducted through the construction of bipartite networks, in which the vertices represent the countries sending tourists and the Brazilian states of destination, while the edges indicate the intensity of the connections between them. Subsequently, the one-mode projection will be applied, allowing the relationships between countries and between states to be analyzed separately. To complement the investigation, histograms will be used—a simple yet effective statistical tool for visualizing variations and changes over the studied period. Additionally, community detection will be performed, a more complex approach aimed at identifying groups of vertices that are more densely connected among themselves than with the rest of the network. This analysis has the potential to reveal interaction patterns that are not perceptible through conventional methods, providing a deeper understanding of the tourism relationships between Europe and Brazil. Thus, this study aims to contribute both to the field of bipartite network analysis and community detection and to the understanding of international tourism in Brazil before, during, and after the COVID-19 pandemic.