Tourism Asset and Spatial Complexity Analyzed Through Graph-Structured Data Analysis

In this work, we aim to move beyond the territorial representation of communities of tourism-related services. Instead, our focus is on exploring the interrelation between two distinct features: tourist attractions and tourism-related services. One approach we consider is to utilize a bipartite graph, a mathematical structure characterized by a division of its vertices into two separate and non-overlapping sets, meaning they have no element in common, such that no two graph vertices within the same set are adjacent. Bipartite networks serve as powerful models for understanding diverse interactions across various disciplines, ranging from social networks to environmental systems. Identifying communities within bipartite networks holds significant importance as it unveils hidden patterns and structures within complex relationships. But instead of relying only on the graph structure, we enhance our understanding of these complex interrelations by integrating Graph Neural Networks (GNNs) into our methodology. GNNs are a type of machine learning model designed specifically for processing input data in the form of graphs. Within these approaches, we can represent a wide range of complex relationships, making them useful for modeling Spatial Interaction in territorial systems, among others.