Sustainability evaluation of basin indicator: A bibliometric analysis using Latent Dirichlet Allocation

This study analyzes the evolution of scientific production related to the assessment of basin sustainability by applying the Latent Dirichlet Allocation~(LDA) model to the content analysis of 2,675 scientific publications. Since 1992, following the United Nations Conference on Environment and Development, the basin has been recognized as the ideal geographic unit for the sustainable management and planning of natural resources. Within this framework, the importance of monitoring basin components has been emphasized in order to evaluate the effectiveness of implemented actions and monitor progress toward sustainable development. The incorporation of tools from Natural Language Processing~(NLP) and Machine Learning~(ML) enabled the identification of the main research topics, as well as thematic and methodological shifts over time. These findings are particularly relevant in the context of the Sustainable Development Goals~(SDGs), as they provide a comprehensive overview of the state of scientific knowledge and its potential contribution to achieving global targets by 2030.