Comparison Of Air Quality Prediction In New Delhi Using Convolutional Neural Network And Multiple Linear Regression Algorithm

This research compares the predictive performance of Convolutional Neural Network (CNN) and Multiple Linear Regression (MLR) algorithms for air quality forecasting in New Delhi. Using a real-world dataset with an 80/20 train-test split, CNN achieved an accuracy of 88%, significantly outperforming MLR’s 59%. Statistical validation using SPSS confirmed the performance gap (p=0.001). The study demonstrates CNN’s superiority for urban air quality prediction and offers insights for future machine learning applications in environmental forecasting.