Unsupervised Density-Based Methods for Interpreting Sustainability Data

The measurement of sustainability across different domains: environmental, economic and social represents a crucial challenge for understanding complex phenomena and supporting responsible development strategies. In this context, density-based clustering algorithms such as DBSCAN (Density- Based Spatial Clustering of Applications with Noise) and HDBSCAN (Hierarchical DBSCAN) offer effective tools for identifying data structures without requiring prior knowledge of the number of clusters. DBSCAN enables the detection of clusters with arbitrary shapes. HDBSCAN, its hierarchical extension, is particularly useful when data exhibit regions of varying density, as is often the case in sustainability analyses. The adoption of these methods allows for the identification of coherent groups of behaviors, territories or entities with shared characteristics, thereby improving the interpretation of indicators and facilitating comparison across different contexts. These approaches have proven valuable, for example, in the analysis of urban policies, corporate practices or environmental scenarios, contributing to a more integrated and comparative understanding of sustainability.