AIR POLLUITON INDEX PREDICTION USING MULTIPLE NEURAL NETWORKS
Air quality monitoring and forecasting tools are necessary for the purpose of taking precautionary measures against air pollution, such as reducing the effect of a predicted air pollution peak on the surrounding population and ecosystem. In this study a single Feed-forward Artificial Neural Network (FANN) is shown to be able to predict the Air Pollution Index (API) with a Mean Squared Error (MSE) and coefficient determination, R2, of 0.1856 and 0.7950 respectively. However, due to the non-robust nature of single FANN, a selective combination of Multiple Neural Networks (MNN) is introduced using backward elimination and a forward selection method. The results show that both selective combination methods can improve the robustness and performance of the API prediction with the MSE and R2 of 0.1614 and 0.8210 respectively. This clearly shows that it is possible to reduce the number of networks combined in MNN for API prediction, without losses of any information in terms of the performance of the final API prediction model.
 ASMA. â€œAir Pollutant Index (API),â€œ Retrieved on July, 2012. Available from http://www.doe.gov.my/portalv1/en/info-umum/english-air-pollutant-index-api/100.
 Akkoyunlu A, Yetilmezsoy K, Erturk F, Oztemel, E. (2010) A neural network-based approach for the prediction of urban SO2 concentrations in the Istanbul metropolitan area. International Journal of Environment and Pollution, 40 (4) : 301-315.
 Wang W, Lu W, Wang X , Leung YT. (2003) Prediction of maximum daily ozone level using combined neural network and statistical characteristics. Environmental International, 29 (5): 555â€“562.
 Viotti P, Liuti G, Di Genova P. (2002) Atmospheric urban pollution: applications of an artificial neural network (ANN) to the city of Perugia. Ecol Modell., 148 (1): 27â€“46.
 Sabri G, Tarek KM. (2012) Combination of artificial neural network models for air quality predictions for the region of Annaba, Algeria. Int. J. Environ. Stud., 69 (1) : 79â€“89.
 Amodio M, Andriani E, Cafagna I, Caselli M, Daresta BE, de Gennaro G, Tutino M. (2010) A statistical investigation about sources of PM in South Italy. Atmos. Res., 98 : 207â€“218.
 Rodriguez S, Querol X, Alastuey A, Kallos G, Kakaliagou O.(2001) Saharan dust contributions to PM10 and TSP levels in Southern and Eastern Spain. Atmos. Environ., 35 :2433â€“2447.
 PeyJ, PÃ©rez N, Querol X, Alastuey A, Cusack M, Reche C. (2010) Intense winter atmospheric pollution episodes affecting the Western Mediterranean. Sci. Total Environ., 408 (8) : 1951â€“1959.
 Pohjola MA, RantamÃ¤ki M, Kukkonen J, Karppinen A, Berge E. (2004) Meteorological evaluation of a severe air pollution episode in Helsinki on 27 â€“ 29 December 1995. Boreal Environ. Res., 9 (1) : 75â€“87.
 De Gennaro G, Trizio L, Di Gilio A, Pey J, PÃ©rez N, Cusack M, Querol X. (2013) Neural network model for the prediction of PM10 daily concentrations in two sites in the Western Mediterranean. Sci. Total Environ., 463-464 : 875â€“883.
 Perez P, Trier A, Reyes J. (2000) Prediction of PM2.5 concentrations several hours in advance using neural networks in Santiago , Chile. Atmos. Environ. , 34 :1189â€“1196.
 Yan CK, Jian L.(2013) Identification of significant factors for air pollution levels using a neural network based knowledge discovery system. Neurocomputing, 99: 564â€“569.
 Gardner MW, Dorling SR. (1998) Artificial neural networks (the multilayer perceptron)â€”a review of applications in the atmospheric sciences. Atmos. Environ., 32(14-15) : 2627â€“2636.
 Perez P, Trier A. (2001) Prediction of NO and NO2 concentrations near a street with heavy traffic in Santiago, Chile. Atmos. Environ., 35: 1783â€“1789.
 Sousa S, Martins F, Alvimferraz M, Pereira M. (2007) Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations. Environ. Model Softw., 22(1) : 97â€“103.
 Ul-Saufie AZ, Yahaya AS, Ramli NA, Rosaida N, Hamid HA. (2013) Future daily PM10 concentrations prediction by combining regression models and feedforward backpropagation models with principle component analysis (PCA). Atmos. Environ., 77 : 621â€“630.
 Cigizoglu KH, Kisi Ã–. (2006) Methods to improve the neural network performance in suspended sediment estimation. J. Hydrol., 317: 221â€“238.
 Chelani AB, Chalapati RC, Phadke K, Hasan M. (2002) Prediction of sulphur dioxide concentration using artificial neural networks. Environ. Model Softw., 17: 161â€“168.
 Gardner MW, Dorling SR. (1999) Neural network modelling and prediction of hourly NO and NO concentrations in urban air in London. Atmos. Environ., 33: 709â€“719.
 Zhang J. (1999) â€œDeveloping Robust Non-linear Models Through Bootstrap Aggregated Neural Networks. Neurocomputing, 25: 93-113.
 Ahmad Z, Zhang J. (2009) Selective combination of multiple neural networks for improving model prediction in nonlinear systems modelling through forward selection and backward elimination. Neurocomputing, 72: 1198-1204.
 Azid A, Juahir H, Latif MT, Mohd Zain S, Osman MR. (2003) Feed-Forward artificial neural network model for Air Pollutant Index prediction in the southern region of Peninsular Malaysia. Journal of Environmental Protection, 4: 1-10.
 Azid A, Juahir H, Toriman ME, Kamarudin MKA, Mohd Saudi AS, ,Che Hasnam CN, Abdul Aziz NA, Zaman F, Latif MT, Mohamed Zainuddin SF, Osman MR, Yamin M.(2014) Prediction of the Level of Air Pollution Using Principal Component Analysis and Artificial Neural Network Techniques: a Case Study in Malaysia. Water Air Soil PolluT., 225, 2063-2077.