Discussion of 'Potential assessment of neural network and decision tree algorithms for forecasting ambient PM_(2.5) and CO concentrations: Case study' by Chandrra Sekar, B. R. Gurjar, C. S. P. Ojha, and Manish Kumar Goyal

摘要

The authors of the original paper presented a case study in which they assessed the potential of artificial neural networks (ANNs) and decision tree algorithms (e.g., REPTree and M5P algorithm techniques) for forecasting ambient PM_(2.5) (particulate matter 2.5 micrometers or less in diameter) and CO concentrations in New Delhi, India. For evaluation of the performances of the models, they used the root mean square error (RMSE), normalized mean square error (NMSE), and Nash-Sutcliffe efficiency index (NSI). They concluded that the M5P algorithm performs better than the ANN and REPTree algorithms, and all the meteorological, traffic, and emission characteristics data are significant in the prediction of PM_(2.5), whereas meteorological and emission input data are significant in the prediction of CO concentrations. The discussers would like to thank the authors for modeling air quality and would like to point out some important points that the authors might like to consider in their future studies.