Air Quality Prediction using Automated ML

Air pollution has grown to be one of the world’s most serious health and environmental issues. The broad range of gases like sulfur dioxide, carbon monoxide, etc which are released into the atmosphere due to the incomplete combustion of fossil fuels results in air pollution. These pollutants must be monitored carefully in order to find out the changes over time and to provide these data for predicting future conditions. Therefore records collected and processed based on those subjects have grown to be an important research location. The main aim of this work is to make use of Automated ML and investigate various machine-learning approaches for air quality forecasting. It makes use of particulate matter using the present and historical values with the combination of values that is readily available. Here, by implementing the proposed parameter-reducing formulation using Automated ML it achieved a better performance than the standard regression models. The proposed model resulted in 80% prediction accuracy when empirically evaluated with UCI data set on pollutant prediction.