In 2008 I was part of a project that studied air quality in Spain and the rest of Europe using self-organizing neural networks. Most of my involvement was concerned with collecting the data from an European Union database in an efficient way, converting it to a format amenable to analysis with neural networks, and performing such an analysis using the SOM Toolbox for Matlab. Our project won the IBM Special Award in the Arquimedes Research Competition for undergraduate researchers, hosted by the Spanish Ministry of Science and Innovation.
A SOM-based methodology for classifying air quality monitoring stationsEnrique Alvarez-Guerra, Abel Molina, Javier R. Viguri, Manuel Alvarez-Guerra
Appeared in Environmental Progress & Sustainable Energy (paywall, don't hesitate to contact me if you have problems accessing the paper)
This article develops a methodology based on the use of Self-Organizing Map (SOM) artificial neural networks for integrating data about multiple measured pollutants to group monitoring stations according to their similar air quality. This methodology is illustrated with its application to a case study in which 517 stations of the Spanish air quality monitoring network were classified considering simultaneously their levels of regulated pollutants in 2005, highlighting some implications of data normalization in the process. Results obtained with the SOM-based methodology, when compared to classifications based directly on legislation, provided more useful classifications for further air quality management actions, and revealed that these types of tools can facilitate the design of air pollution reduction programs by discovering different areas with similar problems.