Modeling of air pollutants using least square support vector regression, multivariate adaptive regression spline, and M5 model tree models

This study investigates the applicability of three different soft computing methods, least square support vector regression (LSSVR), multivariate adaptive regression splines (MARS), and M5 Model Tree (M5-Tree), in forecasting SO2 concentration. These models were applied to monthly data obtained from Janakpuri, Nizamuddin, and Shahzadabad, located in Delhi, India. The models were compared with each other using the cross validation method with respect to root mean square error, mean absolute error, and correlation coefficient. According to the comparison, LSSVR provided better accuracy than the other models, while the MARS model was found to be the second best model in forecasting monthly SO2 concentration. Results indicated that the applied models gave better forecasting accuracy in Janakpuri station than the other stations. The results were also compared with previous studies and satisfactory results were obtained from three methods in modeling SO2 concentrations.

[1]  Osman N. Ucan,et al.  Modeling of SO2 distribution in Istanbul using artificial neural networks , 2005 .

[2]  O. Kisi,et al.  Pan Evaporation Modeling Using Neural Computing Approach for Different Climatic Zones , 2012, Water Resources Management.

[3]  Q Tong,et al.  Measurements and analysis of criteria pollutants in New Delhi, India. , 2001, Environment international.

[4]  Özgür Kisi,et al.  Daily pan evaporation modeling using chi-squared automatic interaction detector, neural networks, classification and regression tree , 2016, Comput. Electron. Agric..

[5]  Kulwinder Singh Parmar,et al.  Statistical variability comparison in MODIS and AERONET derived aerosol optical depth over Indo-Gangetic Plains using time series modeling. , 2016, The Science of the total environment.

[6]  Witold Pedrycz,et al.  Bio-inspired computing for hybrid information technology , 2012, Soft Comput..

[7]  Emilio Corchado,et al.  Soft computing models to identify typical meteorological days , 2011, Log. J. IGPL.

[8]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[9]  Ozgur Kisi,et al.  Fuzzy Genetic Approach for Estimating Reference Evapotranspiration of Turkey: Mediterranean Region , 2013, Water Resources Management.

[10]  Kulwinder Singh Parmar,et al.  River Water Prediction Modeling Using Neural Networks, Fuzzy and Wavelet Coupled Model , 2014, Water Resources Management.

[11]  Jan Adamowski,et al.  Modeling of daily pan evaporation in sub tropical climates using ANN, LS-SVR, Fuzzy Logic, and ANFIS , 2014, Expert Syst. Appl..

[12]  Baridalyne Nongkynrih,et al.  “Air pollution in Delhi: Its Magnitude and Effects on Health” , 2013, Indian journal of community medicine : official publication of Indian Association of Preventive & Social Medicine.

[13]  Kirti Soni,et al.  Clear-sky direct aerosol radiative forcing variations over mega-city Delhi , 2010 .

[14]  Davor Z Antanasijević,et al.  PM(10) emission forecasting using artificial neural networks and genetic algorithm input variable optimization. , 2013, The Science of the total environment.

[15]  Ozgur Kisi,et al.  Fuzzy Genetic Approach for Modeling Reference Evapotranspiration , 2010 .

[16]  R. B. Singh,et al.  Aerosols over Delhi during pre‐monsoon months: Characteristics and effects on surface radiation forcing , 2005 .

[17]  J. Kukkonen,et al.  Intercomparison of air quality data using principal component analysis, and forecasting of PM₁₀ and PM₂.₅ concentrations using artificial neural networks, in Thessaloniki and Helsinki. , 2011, The Science of the total environment.

[18]  Ozgur Kisi,et al.  Lake Level Forecasting Using Wavelet-SVR, Wavelet-ANFIS and Wavelet-ARMA Conjunction Models , 2015, Water Resources Management.

[19]  Peter S. Sephton,et al.  Forecasting recessions: can we do better on MARS? , 2001 .

[20]  A. Etemad-Shahidi,et al.  COMPARISON BETWEEN M5 MODEL TREE AND NEURAL NETWORKS FOR PREDICTION OF SIGNIFICANT WAVE HEIGHT IN LAKE SUPERIOR , 2009 .

[21]  O. Kisi.  Pan evaporation modeling using least square support vector machine, multivariate adaptive regression splines and M5 model tree , 2015 .

[22]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[23]  Osman N. Ucan,et al.  A New Approach to Prediction of SO2 and PM10 Concentrations in Istanbul, Turkey: Cellular Neural Network (CNN) , 2011 .

[24]  J. Seinfeld,et al.  Atmospheric Chemistry and Physics: From Air Pollution to Climate Change , 1997 .

[25]  Ozgur Kisi,et al.  Modeling monthly pan evaporations using fuzzy genetic approach , 2013 .

[26]  David G. Streets,et al.  Aura OMI observations of regional SO2 and NO2 pollution changes from 2005 to 2015 , 2015 .

[27]  J. Friedman.  Multivariate adaptive regression splines , 1990 .

[28]  Ercan Oztemel,et al.  A neural network-based approach for the prediction of urban SO2 concentrations in the Istanbul metropolitan area , 2010 .

[29]  Ozgur Kisi,et al.  Daily pan evaporation modelling using multi‐layer perceptrons and radial basis neural networks , 2009 .

[30]  Dimitri P. Solomatine,et al.  M5 Model Trees and Neural Networks: Application to Flood Forecasting in the Upper Reach of the Huai River in China , 2004 .

[31]  Anup K. Prasad,et al.  Aerosol radiative forcing over the Indo-Gangetic plains during major dust storms , 2007 .

[32]  N. Pérez,et al.  Neural network model for the prediction of PM10 daily concentrations in two sites in the Western Mediterranean. , 2013, The Science of the total environment.

[33]  Álvaro Herrero,et al.  Neural visualization of network traffic data for intrusion detection , 2011, Appl. Soft Comput..

[34]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[35]  Achuthan Jayaraman,et al.  Wintertime aerosol properties during foggy and nonfoggy days over urban center Delhi and their implications for shortwave radiative forcing , 2006 .

[36]  P. Goyal,et al.  Present scenario of air quality in Delhi: a case study of CNG implementation , 2003 .

[37]  Johan A. K. Suykens,et al.  Support Vector Machines: A Nonlinear Modelling and Control Perspective , 2001, Eur. J. Control.

[38]  B. R. Gurjar,et al.  Emission Estimates and Trends (1990-2000) for Megacity Delhi and Implications , 2004 .

[39]  Aytac Guven,et al.  Daily pan evaporation modeling using linear genetic programming technique , 2011, Irrigation Science.

[40]  O. Kisi,et al.  Application of least square support vector machine and multivariate adaptive regression spline models in long term prediction of river water pollution , 2016 .

[41]  Shiv O. Prasher,et al.  APPLICATION OF MARS IN SIMULATING PESTICIDE CONCENTRATIONS IN SOIL , 2006 .

[42]  Francisco Javier de Cos Juez,et al.  Bankruptcy forecasting: A hybrid approach using Fuzzy c-means clustering and Multivariate Adaptive Regression Splines (MARS) , 2011, Expert Syst. Appl..

[43]  Anurag Kandya,et al.  An Analysis of the Annual and Seasonal Trends of Air Quality Index of Delhi , 2007, Environmental monitoring and assessment.

[44]  Ozgur Kisi,et al.  Predicting daily pan evaporation by soft computing models with limited climatic data , 2015 .

[45]  Mahesh Pal,et al.  M5 model tree based modelling of reference evapotranspiration , 2009 .

[46]  Yong Liu,et al.  A novel hybrid forecasting model for PM₁₀ and SO₂ daily concentrations. , 2015, The Science of the total environment.

[47]  Kulwinder Singh Parmar,et al.  Statistical analysis of aerosols over the Gangetic–Himalayan region using ARIMA model based on long-term MODIS observations , 2014 .

[48]  B. R. Gurjar,et al.  Human health risks in megacities due to air pollution , 2010 .

[49]  Kulwinder Singh Parmar,et al.  Time series model prediction and trend variability of aerosol optical depth over coal mines in India , 2015, Environmental Science and Pollution Research.