Data quality is one of the fundamental issues influencing the performance of any data investigation algorithm. Poor data quality always leads to poor quality results. In the investigation chain, the data selection phase is followed by the preprocessing phase, which results in increased data quality, while in parallel it demands the highest time resources of the overall data investigation chain. The preprocessing phase includes the handling of missing data, handling of the outliers, data de-trending and data smoothing. The methods that are used in the preprocessing phase are usually not sufficiently reported in the literature of environmental data analysis and knowledge extraction. The current paper investigates the performance of several methods in all phases of the preprocessing chain of environmental data, by emphasizing in the use of ICT (Information & Communication Technology) methods for the materialization of such preprocessing tasks, and by making use of the air quality as the environmental domain paradigm. © Springer-Verlag Berlin Heidelberg 2009.
CITATION STYLE
Kyriakidis, I., Karatzas, K. D., & Papadourakis, G. (2009). Using preprocessing techniques in air quality forecasting with artificial neural networks. Environmental Science and Engineering (Subseries: Environmental Science), 357–372. https://doi.org/10.1007/978-3-540-88351-7_27
Mendeley helps you to discover research relevant for your work.