Prediction of Air Quality using Deep Learning based Model with CRO

As their populations rapidly urbanise, many emerging nations also have a severe issue with air pollution. The ability to foresee changes in air quality is increasingly valuable to both policymakers and citizens. In this study, we usage of air quality and weather forecast data to make predictions for the following 48 hours at each monitoring station. This work creates a feature selection-based classifier for forecasting based on expertise in the field of air pollution. Three forms of information-air contaminants, meteorological data, and weather data-are offered as prerequisites for making air quality forecasts. In order to facilitate more accurate categorization, the raw data are split and standardised during the pre-processing stage. Therefore, it is essential to identify the best characteristics from the pre-processed data and eliminate the extraneous ones that might lead to incorrect categorization. In this research, an unique optimization-based approach was used to pick features from coral reefs. An optimization approach called the coral reefs optimization (CRO) procedure is able to find optimal solutions by modelling coral behaviours relevant to reef location and development. Each possible solution to the issue is analogized to a coral that is always searching for a suitable spot in which to settle and flourish in the reefs in the suggested approach. At each stage, the solutions are processed with the help of the coral reefs optimization algorithm's unique operators. A better solution is more likely to flourish on the reefs after a number of iterations. The issue is solved by selecting the best solution. In the end, a Network is used to make predictions (DNN). As can be seen from the findings, the suggested model is almost as accurate as the state-of-the-art alternatives, at 96%.