Machine learning-based detection of sudden air pollutant level changes: impacts on public health

Pollution originating from human activities can harm both humans and ecosystems. Our research underscores the detrimental impact of human-induced pollution on both human health and ecosystems. Prolonged exposure to pollution poses health risks, but sudden surges in pollutant levels can be even more hazardous. We have devised a robust method to detect abrupt changes in air quality resulting from human activities. Our study involved real-time analysis of PM \(_{2.5}\) ${}_{2.5}$ and PM \(_{10}\) ${}_{10}$ pollution data collected over 18 months from an industrial-based suburban city. We meticulously analyzed spatial and meteorological patterns in pollutant concentrations using a grid-based system equipped with monitoring devices. Remarkably, our approach exhibited an impressive 97.83% accuracy in predicting sudden air quality changes. By comprehensively understanding pollution patterns, our research contributes significantly to the development of enhanced remediation techniques.