Magnitudes of the air pollution depend on various variables. Preprocessing and optimisation processes are necessary to discover the complexity of the relationship of the data for more accurate and efficient predictions. These techniques help to clean the datasets and to find the best structure of the smooth data model. The Gamma test (GT) and Genetic Algorithm (GA) are practical tools which can be applied for preprocessing and optimising data models. Regarding building a smooth data model, the developed artificial neural networks are functional optimisation strategies which are suitable for ANN training. Moreover, local linear regression (LLR) and dynamic local linear regression (DLLR) models are effective due to the high density of our normalised dataset. In this regard, we developed a process to construct a smooth data model to support environmental decision making in air pollution emergency conditions. The main objective of this work was to set an appropriate algorithm by preprocessing and optimising a set of the data model for developing smooth data-driven models which could play a significant role in early warning systems in regard to the human health. The data sets included the meteorological and air pollutant variables as inputs/predictors and emergency medical service clients as outputs. The GT and GA were applied to analyse and optimise the input variables. Three types of ANNS (ANN1, ANN2, and ANN3), (LLR), and (DLLR) techniques were used to establish the models. Finally, a smooth data model was constructed and evaluated. |
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Published by: National Cave Research and Protection Organization, India
<Environmental Science+Zoology+Geology+Cave Science>AMBIENT SCIENCE