Abstract
Artificial neural network-based air pollution prediction models have become very popular. The paper describes feature determination and pattern selection strategies that help to improve significantly the performance of neural-network based models.
The most important methods of feature determination are preprocessing, heuristic determination, feature extraction and feature selection. The most important methods for pattern selection are meteorological knowledge-based cluster determination and Kohonen neural network-based cluster determination.
To summarize the explanation of the proposed methods, their influence on model behaviour is discussed.
Keywords: Artificial neural networks, air pollution, prediction model, feature determination, pattern selection, cluster.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Beleue, L. M., Bauer, K. W. Jr., 1995, Determing input features for multilayer perceptrons, Neurocomputing, vol. 7, 2, 111–121
Božnar, M., Lesjak, M., Mlakar, P., 1993, A neuralnetwork-based method for short-term predictions of ambient SO2 concentrations in highly polluted industrial areas of complex terrain. Atmos. environ., B Urban atmos., vol. 27, 221–230.
Božnar, M., Brusasca, G., Cavicchioli, C., Faggian, P., Finardi, S., Mlakar, P., Morselli, M. G., Sozzi, R., Tinarelli, G., 1994, Application of advanced and traditional diffusion models to en experimental campaign in complex terrain. V: Baldasano, J. M., ed., Air Pollution II. Volume 1, Computer simulation. Southampton; Boston: Computational Mechanics Publications, 159–166.
Božnar, M., Mlakar, P., 1995, Neural networks — a new mathematical tool for air pollution modelling. V: Power, H. ed., Moussiopoulos, N. ed., Brebbia, C. A., ed., Air pollution III. Volume 1, Air pollution theory and simulation. Southampton; Boston: Computational Publications, 259–266.
Božnar, M., Mlakar, P., 1996, Analysis of ambient SO2 concentrations in the surroundings of the šoštanj thermal power plant. V: Caussade, B. ed., Power, H. ed., Brebbia, C. A., ed., Air pollution IV: monitoring, simulation and control. Southampton; Boston: ComputationalMechanics Publications, 727–736.
Božnar, M., Mlakar, P., 19971, Pattern selection strategies for SO2 forecasting models. V: Power, H. (Ur.), Tirabassi, T. ed., Brebbia, C. A., Ed., Air pollution V: Šmodelling,monitoring and management]. Southampton; Boston: Computational Mechanics Publications, 547–556.
Bonar, M., 19972, Pattern selection strategies for a neural network — based short term air pollution prediction model. V: ADELI, H., ed., Intelligent Information Systems IIS’97, Grand Bahama Island, Bahamas, December 8–10, 1997. Proceedings. Los Alamitos, California: IEEE Computer Society, 340–344
Božnar, M., Mlakar, P., 1998, Improvement of air pollution forecasting models using feature determination and pattern selection strategies. V: Gryning,. Sven-Erik,. ed., Chaumerliac, Nadine. Air pollution modeling and its application XII, (NATO challenges of modern society, vol. 22). New York; London: Plenum Press, 725–726.
Cybenko, G., 1989, Approximation by superposition of a sigmoidal function, Mathematics of control, signals and systems, 2, 303–314
Kohonen, T., 1995, Self-organising maps, Springer-Verlag, Berlin
Mlakar, P., Božnar, M., 19941, Short-term air pollution prediction on the basis of artificial neural networks. V: Baldasano, J. M., ed., Air Pollution II. Volume 1, Computer simulation. Southampton; Boston: Computational Mechanics Publications, 545–552.
Mlakar, P., Božnar, M., Lesjak, M., 19942, Neural networks predict pollution. V:GRYNING, Swen-Erik. ed., MILLÁN, Millán M., ed., Air pollution modeling and its applications (NATO challenges of modern society, 18). New York; London: Plenum Press, 531–532
Mlakar, P., Božnar, M., 1996, Analysis of winds and SO2 concentrations in complex terrain. V: Caussade, B. ed., Power, H. ed., Brebbia, C. A., ed., Air pollution IV: monitoring, simulation and control. Southampton; Boston: Computational Mechanics Publications, 455–464.
Mlakar, P., Božnar, M., 19971, Feature determination for air pollution forecasting models. V: POWER, H. ed., Tirabassi, T. (Ur.), Brebbia, C. A., ed., Air pollution V: Šmodelling, monitoring and management]. Southampton; Boston: Computational Mechanics Publications, 577–586.
Mlakar, P., Božnar, M., 19972, Perceptron neural network — based model predicts air pollution. V: Adeli, H., ed., IntelligentInformationSystems IIS’97, Grand BahamaIsland, Bahamas,December 8–10, Proceedings. Los Alamitos, California: IEEE Computer Society, 345–349.
Mlakar, P., 19973, Determination of features for air pollution forecasting models, V: Adeli, H., ed., Intelligent Information Systems IIS’97, Grand Bahama Island, Bahamas, December 8–10, Proceedings. Los Alamitos, California: IEEE Computer Society, 350–354.
Narendra, P. M., Fukunaga, K., 1977, A branch and boudn algorithm for Feature subset selection, IEE trans. Comput., vol. 26, 917–922
Ruck, D. W., Rogers, S. K., Kabrisky, M., 1990, Features selection using multilayer perceptron, Neural network comput. 20, 40–48
Rumelhart, D. E., McClelland, J. L., 1986, Parallel distributed processing 1,2, MIT Press, Cambridge, MA
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Kluwer Academic Publishers
About this chapter
Cite this chapter
Božnar, M.Z., Mlakar, P. (2004). Artificial Neural Network-Based Environmental Models. In: Gryning, SE., Schiermeier, F.A. (eds) Air Pollution Modeling and Its Application XIV. Springer, Boston, MA. https://doi.org/10.1007/0-306-47460-3_49
Download citation
DOI: https://doi.org/10.1007/0-306-47460-3_49
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-306-46534-5
Online ISBN: 978-0-306-47460-6
eBook Packages: Springer Book Archive