Smart City Air Quality Management with IoT and Bayesian Optimization for Pollution Monitoring

The rapid urbanization happening around the globe is having a huge effect on the environment. Cities in the poor world are particularly vulnerable to air pollution. In light of this issue, several nations are mandating that cities implement plans to enhance air quality, and the new global air quality recommendations from the World Health Organization (WHO) are adding fuel to the fire. Inadequate outreach, few observations, disconnected city operations, and inconsistent protocols are some of the issues that hinder these deployments, as does the absence of collaborative UAQM governance. The proposed approach consists of three phases, which are data preparation, feature selection, and training. When it comes to categorical preprocessing, there are two typical ways to deal with missing values are after removing rows with missing values, use KNN Imputer to fill in the blanks. One goal of feature selection methods is to make analysis easier and faster by reducing the target dataset's dimensionality. Training the model was done using BO. With an average accuracy rate of 93.14 percent, the proposed model beats out the alternatives, including SVM and RBF.