Development of artificial intelligence based NO 2 forecasting models at Taj Mahal, Agra

A novel artificial intelligence-based model outperforms traditional methods for forecasting NO₂ concentrations at the Taj Mahal, Agra. The PCA–ANN (Principal Component Analysis–Artificial Neural Network) model demonstrated superior accuracy with a correlation coefficient of 0.91 during validation, compared to multiple linear regression (MLR), which achieved 0.89. This study highlights the potential of computational intelligence in public health-oriented air quality forecasting, particularly for high-pollution periods like November when levels spike significantly around the monument. The model's reliability was validated through statistical metrics including NMSE, IOA, and FB, all showing strong agreement with observed data.

Mishra, D., & Goyal, P. (2015). Development of artificial intelligence based NO 2 forecasting models at Taj Mahal, Agra. Atmospheric Pollution Research, 6(1), 99–106. https://doi.org/10.5094/apr.2015.012