Developing an Air Quality Index Model Predictor

Abstract

Pollutants and particles have a dynamic character, are highly volatile, and exhibit a high degree of temporal and spatial fluctuation, making it challenging to forecast air quality accurately. On the other hand, the ability to model, forecast, and monitor air quality is becoming more critical, particularly in metropolitan areas. This is because air pollution has significantly impacted the environment and human health. In this paper, we employ extra tree, random forest, Linear Discriminant Analysis, K Neighbors, Logistic regression, and ensemble [random forest, extra tree] machine learning methods to classify the air quality of India from 2015 to 2020. The presented results demonstrate that ensemble [random forest, extra tree] allows us to classify daily AQI (Air Quality Index) for all India's cities accurately.