At the beginning of 2019, a gasoline supply crisis severely disrupted Mexico’s economy, which heavily relies on the petroleum industry. In response, the Mexican government launched initiatives to identify locations and officials involved in gasoline theft from PEMEX (Petróleos Mexicanos), the country’s state-owned oil company. These efforts included monitoring the national pipeline network to detect theft points. However, these measures resulted in severe gasoline shortages in major urban areas, prompting widespread reactions on social media and generating open data from news outlets. Despite the government’s release of open data on gasoline theft locations, no comprehensive analysis has been conducted to explore patterns, insights, and the spatio-temporal dynamics of this crisis. This paper presents a case study analyzing social media reactions and their correlation with government open data during the gasoline supply crisis in Mexico’s urban population. We propose a framework to uncover trends and public perceptions emerging from social media and their relationship with official data. Utilizing machine learning techniques based on linguistic and semantic analysis, we examined a dataset of 24,317 geo-referenced tweets. The findings reveal public opinion dynamics, polarization patterns, and regional insights. A key trend identified was the prevalence of long fuel lines, ranging from 1.5 to 5 km, at fuel stations across multiple Mexican states.