Análisis de la contaminación del aire por material particulado usando minería de texto y análisis de series temporales en Colombia

Abstract

Every year the loss of life due to air pollution is recorded, and in this regard the different entities have developed actions that involve from the formulation of the regulation to the implementation of decontamination and health plans. In Colombia, this figure corresponds to about 8000 deaths annually that are related to diseases of the respiratory and cardiovascular systems, influenced by exposure to atmospheric pollutants. However, related information is not available to the entire community; Although the information may be concentrated in technical reports or at the institutional level, its reach to the entire community is limited. It is also found that in monitoring stations in Medellín in the jurisdiction of the Valle de Aburra Metropolitan Area (AMVA) and in Bogotá, in the jurisdiction of the District Department of the Environment (SDA), the highest concentrations of particulate matter in the country have been reported. , at different time periods in the last ten years. In this order of ideas, the need to identify the degree of disclosure to the community through the mass media, such as newspaper articles, and its relationship with the behavior of the particulate matter air pollutant in its PM10 and PM2 fraction is observed. 5. For this purpose, it is necessary to apply text mining tools, time series analysis and machine learning to identify the frequency of terms, thematic grouping, the possible relationship between the published information and the behavior of air pollutants registered by the stations. that monitor the air quality in the jurisdiction of the AMVA and the SDA with the greatest non-compliance with the norm registered in the last ten years. This project is expected to identify the aspects required in terms of disclosure to the community, as well as the most frequently disclosed information in support of the implementation of measures that support the prevention and mitigation of the effects of exposure to particulate matter and contribute to the achievement of SDG 11 in its air quality component