A disaster transforms the reality of those affected, as it results in the loss of homes, schools, jobs, biodiversity, and non-material assets. The aspects influenced by a disaster make up the Sustainable Development Goals (SDG), seventeen global challenges that must be overcome by 2030. The progress of compliance with the SDG can be monitored by indicators that reflect the objectives according to the specificities of each country. Due to the social, economic, and environmental correlation between SDG and disasters, it is important to consider computational strategies, such as Machine Learning (ML), that enable the analysis of these two scenarios. In this article, we describe and compare the performance and difficulties of six ML experiments to generate fuzzy rules that enable the analysis of the relationship between disasters in Brazil and social and economic indicators that can be further related to SDG. To this end, we used the Fuzzy Association Rule-Based Classification Model (FARC-HD). There are contributions in the literature regarding ML models used in the disaster scenario, but the interrelationship between the SDG and events that affected Brazil has not yet been explored. The results indicated a preliminary set of indicators that can be used to associate SDG 1 and 10 with disasters. Although this article focuses on the Brazilian context, the methodologies and considerations presented can be extended to other scopes.