A Machine Learning Model for Air Quality Prediction in Smart Cities

Air pollution is one of the major environmental issues facing our planet and contributing to its global warming. As defined by the world health organization (WHO), air pollution denotes the presence of harmful substances that cause detrimental changes to air quality. Numerous studies have shown that long-term exposure to such toxic and poisonous air components has several adverse effects on human health and the living ecosystem in general. Furthermore, statistical studies have shown that pollution is tightly associated with mortality and morbidity. This is because it can cause a variety of diseases that range from cough and asthma to heart attacks and cancers. The most common air pollutants can be broadly classified into three categories: solid particles (e.g. mercury, lead), gaseous pollutants ( Nitrogen oxide, NO2, Carbon monoxide, CO, sulphur dioxide, SO2), and a mixture of liquid droplets and small particles called particulate matters .The concentrations of such components define the quality of the air. At high air pollutants’ concentrations, the atmospheric composition experiences severe changes, and is thus subject to a higher damage. Therefore, pollution levels can be evaluated by measuring the concentration of such elements in the air. These measurements are generally performed by air monitoring stations on hourly, daily, and monthly basis.