An Asynchronous Cellular Automata-based Multi-class Pattern Classification Model for Air Quality Index Prediction

With the rapid growth of industrialization, significant changes in the composition of ambient air have been observed, leading to increasing concerns about air pollution. The concentration of air pollutants is commonly represented using a statistical measure known as the Air Quality Index (AQI), which indicates the level of air pollution and its potential adverse effects on human health. In this study, AQI prediction for effective air pollution monitoring in Durgapur city is investigated. The primary objective of this work is to develop an efficient asynchronous cellular automata (ACA)-based model for classifying air pollutant patterns and predicting AQI under both null and periodic boundary conditions. Experimental results demonstrate that the proposed ACA-based approach performs comparably to existing machine learning-based methods for AQI prediction. Moreover, the proposed ACA-based classifier achieves maximum testing accuracies of 88.11% under the null boundary configuration and 91.83% under the periodic boundary configuration, outperforming several existing classification approaches.