AI Tool for Personalised Recommended System on Forecasting Air Pollutants

Air pollution is one of the worst enemies of public health, as it leads to respiratory and cardiovascular diseases while reducing average life expectancy. Most air-quality monitoring systems existing today provide only real-time values of pollution and cannot reliably forecast future concentration, thus failing to give advance warning to the people. This leads to a requirement of a system which assists individuals in predicting risks of ill health due to low air quality. This paper, therefore, introduces an AI-based approach to forecasting air pollution that calculates the levels of various pollutants for a future time by using historical air-quality records and important meteorological variables such as temperature, humidity, and wind speed. Linear regression and random forest algorithms examine existing data to make predictions of pollution levels on the meteorological parameter. The results obtained indicate that accurate predictions are offered with these models and that particulate matter, especially Particulate matter2.5 and Particulate matter10, is an important factor in poor air quality. Based on the levels of predicted pollution, the system provides general health advisories which apprise users of possible risks, including respiratory or cardiovascular discomfort during highpollution conditions. This approach supports the creation of early awareness and encourages precautionary actions in order to reduce the threat of health risks related to exposure.