Predict Air Pollution With AI

A new technology to predict air pollution levels in advance has been developed by computer scientists at Loughborough University. It uses artificial intelligence (AI) to predict particulate matter of less than 2.5 microns in diameter. Presence of particulate matter (PM2.5) reduces visibility and creates haziness in cities. Being too small, the particulate matter can easily enter the lungs and then the bloodstream resulting in cardiovascular, cerebrovascular and respiratory impacts. The system is capable of predicting from one hour to several hours or even upto two days in advance, the causes for the increase in PM2.5 including weather, seasonal and environmental factors and it can also be used as a tool for air pollution analysis in a carbon credit trading system. The system’s uncertainty analysis and ability to understand factors that affect PM2.5 are particularly important as this will allow potential end-users, policymakers and scientists to better understand related causes of PM2.5 and how reliable the prediction is.