An extensive survey on machine and deep learning algorithms for air quality analysis

Imbalance of dust in air produce unfavorable consequences on the living organisms is called as air pollution. Air pollution affects entire Environment significantly. The main source of air pollution is Pollutant. Large amount of toxic or harmful pollutants such as SO2, NO2, CO, Particulate Matter (PM) of various sizes(10µm,2.5µm) are emitting from smoke exhaust from industries, power plant emission, vehicle emission, burning of fossil fuels. The pollution may leads to serious health issues to the people which affects an entire environment dramatically. Particulate Matter leads to greenhouse gases and global warming. For the earlier period, some of the approaches are preferred to find out the quality of the air, AI is the most generally used approach to find out the quality of air, here by some of the sub set techniques are also used like machine learning and deep learning approach. A Neural network act as the heart of deep learning. It carries artificial neural networks one step beyond with using huge real time data set, providing more trustable results. In reality, air pollution could affect quickly our health, so it is important to have real time air pollution tracking data to make an early prediction and alerting. The large amount of historical data available for analysis and the need for performing more accurate forecasts in pollutant areas motivates to enhance the process with the help of emerging technologies like deep learning approaches. The core of this study is to learn the process of various approaches and models available in existing survey for the detection of air pollutant present in air and to develop a hybrid intelligent model for concentration estimation of various pollutants present in the air along with weather features. In reality, air pollution could affect quickly our health, so it is important to have real-time air pollution tracking data to make an early prediction and alerting