Abstract
The main aim of this paper is to predict NO and NO2 concentrations 4 days in advance by comparing two artificial intelligence learning methods, namely, multi-layer perceptron and support vector machines, on two kinds of spatial embedding of the temporal time series. Hourly values of NO and NO2 concentrations, as well as meteorological variables were recorded in a cross-road monitoring station with heavy traffic in Szeged, in order to build a model for predicting NO and NO2 concentrations several hours in advance. The prediction of NO and NO2 concentrations was performed partly on the basis of their past values, and partly on the basis of temperature, humidity and wind speed data. Since NO can be predicted more accurately, its values were considered primarily when forecasting NO2. Time series prediction can be interpreted in a way that is suitable for artificial intelligence learning. Two effective learning methods, namely, multi-layer perceptron and support vector regression are used to provide efficient non-linear models for NO and NO2 time series predictions. Multi-layer perceptron is widely used to predict these time series, but support vector regression has not yet been applied for predicting NO and NO2 concentrations. Three commonly used linear algorithms were considered as references: 1-day persistence, average of several day persistence and linear regression. Based on the good results of the average of several day persistence, a prediction scheme was introduced, which forms weighted averages instead of simple ones. The optimization of these weights was performed with linear regression in linear case and with the learning methods mentioned in non-linear case. Concerning the NO predictions, the non-linear learning methods give significantly better predictions than the reference linear methods. In the case of NO2, the improvement of the prediction is considerable, however, it is less notable than for NO.
Similar content being viewed by others
References
Gardner MW, Dorling SR (1998) Artificial neural networks (the multi-layer perceptron)—a review of applications in the atmospheric sciences. Atmos Environ 32:2627–2636
Gardner MW, Dorling SR (1999) Neural network modelling and prediction of hourly NOx and NO2 concentrations in urban air in London. Atmos Environ 33:709–719
Jorquera H, Pérez R, Cipriano A, Espejo A, Letelier MV, Acuňa G (1998) Forecasting ozone daily maximum levels at Santiago, Chile. Atmos Environ 32:3415–3424
Perez P, Trier A, Reyes J, (2000) Prediction of PM2.5 concentrations several hours in advance using neural networks in Santiago, Chile. Atmos Environ 34:1189–1196
Perez P, Trier A (2001) Prediction of NO and NO2 concentrations near a street with heavy traffic in Santiago, Chile. Atmos Environ 35:1783–1789
Perez P, (2001) Prediction of sulfur dioxide concentrations at a site near downtown Santiago, Chile. Atmos Environ 35:4929–4935
Perez P, Reyes J (2001) Prediction of particlulate air pollution using neural techniques. Neural Comput Appl 10(2):165–171
Chelani AB, Chalapati RCV, Phadke KM, Hasan MZ (2002) Prediction of sulphur dioxide concentration using artificial neural networks. Environ Modell Softw 17(2):159–166
Mechaqrane A, Zouak M (2004) A comparison of linear and neural network ARX models applied to a prediction of the indoor temperature of a building. Neural Comput Appl 13(1):32–37
Maqsood I, Riaz Khan M, Ajith Abraham A (2004) An ensemble of neural networks for weather forecasting. Neural Comput Appl 13(2):112–122
Agirre-Basurko E, Ibarra-Berastegib G, Madariaga I (2006) Regression and multilayer perceptron-based models to forecast hourly O3 and NO2 levels in the Bilbao area. Environ Modell Softw 21(4):430–446
Hansen JV, McDonald JB, Nelson RD (1999) Time series prediction with genetic-algorithm designed neural networks: an empirical comparison with modern statistical models. Comput Intell 15:171–184
Small M, Tse CK (2002) Minimum description length neural networks for time series prediction. Phys Rev E 66:066701-1–066701-12
Castillo O, Melin P (2002) Hybrid intelligent systems for time series prediction using neural networks, fuzzy logic and fractal theory. IEEE T Neural Netw 13:1395–1408
Kukkonen J, Partanen L, Karppinen A, Ruuskanen J, Junninen H, Kolehmainen M, Niska H, Dorling S, Chatterton T, Foxall R, Cawley G (2003) Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurements in central Helsinki. Atmos Environ 37:4539–4550
Ordieres JB, Vergara EP, Capuz RS, Salazar RE (2005) Neural network prediction model for fine particulate matter (PM2.5) on the US–Mexico border in El Paso (Texas) and Ciudad Juárez (Chihuahua). Environ Modell Softw 20(5):547–559
Blum E, Li L (1991) Approximation theory and feedforward networks. Neural Netw 4:511–515
Chester D (1990) Why two hidden layers are better than one. In: Erlbaum L (ed) International Joint Conference on Neural Networks, Proceedings 1, Washington, D.C., pp 265–268
Hornik K (1991) Approximation capabilities of multi-layer feedforward networks. Neural Netw 4(2):251–257
Drucker H, Burges CJC, Kaufman L, Smola A, Vapnik V (1997) Support vector regression machines. In: Mozer M et al (eds.). Advances in neural information processing systems, 9, The MIT Press, Cambridge, pp 155–161
Suykens JAK, De Brabanter J, Lukas L, Vandewalle J (2002) Weighted least squares support vector machines: robustness and sparse approximation. Neurocomputing 48:85–105
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Disc 2(2):121–167
Shawe-Taylor J, Cristianini N (2004) Kernel methods for pattern analysis. Cambridge University Press
Vapnik V, Golowich S, Smola A (1997) Support vector method for function approximation, regression estimation, and signal processing. In: Mozer M, et al (eds) Advances in neural information processing systems 9. The MIT Press, Cambridge, pp 281–287
Schölkopf B, Bartlett P, Smola A, Williamson R (1998) Support vector regression with automatic accuracy control. In: Niklasson L et al (eds) Proceedings of the international conference on artificial neural networks, perspectives in neural computing. Springer, Berlin, pp 111–116
Ancona N (1999) Properties of support vector machines for regression. Technical report, Instituto Elaborazione segnali ed immagini, Bari, Italy, pp. 01–99
Müller KR, Smola A, Rätsch G, Schölkopf B, Kohlmorgen J, Vapnik V (1999) Using support vector machines for time series prediction. In: Schölkopf B et al (eds) Advances in kernel methods—support vector learning. Proceedings of the NIPS, workshop on support vectors. The MIT Press, Cambridge, pp 1–12
Schölkopf B, Burges CJC, Smola AJ (eds) (1999) Advances in kernel methods—support vector learning, proceedings of the NIPS workshop on support vectors. The MIT Press, Cambridge
Van Gestel T, Suykens J, Baestaens D, Lambrechts A, Lanckriet G, Vandaele B, De Moor B, Vandewalle J (2001) Financial time series prediction using least squares support vector machines within the evidence framework. IEEE T Neural Network, Spec Issue Neural Network Financ Eng 12(4):809–821
Reed RD, Marks RJ (1999) Neural smithing: supervised learning in feedforward artificial neural networks. MIT Press, Cambridge
Schölkopf B (1997) Support Vector Learning. PhD Thesis. Oldenbourg R. Verlag, Munich
Lin H, Lin C (2003) A study on sigmoid kernels for SVM and the training of non-PSD kernels by SMO-type methods. Technical report, department of computer science and information engineering, National Taiwan University
Kecman V (2001) Learning and soft computing: support vector machines, neural networks, and fuzzy logic models. The MIT Press, Cambridge
Péczel, G (1979) Climatology (in Hungarian), Tankönyvkiadó, Budapest, pp. 258–284
Weigend AS, Gershenfeld NA (eds) (1994) Time series prediction: forecasting the future and understanding the past. Addison–Wesley, Reading, MA
Hammer B, Gersmann K (2003) A note on the universal approximation capability of support vector machines. Neural Process Lett 17:43–53
Witten IH, Frank E (2000) Data mining: practical machine learning tools with Java implementations, Morgan Kaufmann, San Francisco
Weka data mining software library. http://www.cs.waikato.ac.nz/ml/weka. Accessed February 2005
Chang CC, Lin CJ (2002) Training nu-support vector regression: theory and algorithms, Neural Comput 14(8):1959–1977
LibSVM support vector software library http://www.csie.ntu.edu.tw/~cjlin/libsvm . Accessed February 2005)
Makra L, Mayer H, Mika J, Sánta T, Holst J (2008) Variations of traffic related air pollution on different time scales in Szeged, Hungary and Freiburg, Germany. Phys Chem Earth (accepted)
Shi P, Harrison RM (1997) Regression modelling of hourly NO x and NO2 concentrations in urban air in London. Atmos Environ 31(24):4081–4094
Acknowledgments
The authors are indebted to Gábor Motika (Environmental protection inspectorate of Lower-Tisza Region, Szeged, Hungary) for handing monitoring data of the meteorological parameters and the air pollutants and Zoltán Sümeghy (Department of climatology and landscape ecology, University of Szeged, Hungary) for digital mapping, as well as Rita Béczi (Department of climatology and landscape ecology, University of Szeged, Hungary) for useful suggestions. This study was supported by the EU-6 Project “QUANTIFY” [No. 003893 (GOCE)] and the high performance computing group of the University of Szeged.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Juhos, I., Makra, L. & Tóth, B. The behaviour of the multi-layer perceptron and the support vector regression learning methods in the prediction of NO and NO2 concentrations in Szeged, Hungary. Neural Comput & Applic 18, 193–205 (2009). https://doi.org/10.1007/s00521-007-0171-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-007-0171-1