Self-Organizing Maps (SOMs), or Kohonen networks, are widely used neural network architecture. This paper starts with a brief overview of how SOMs can be used in different types of problems. A simple and intuitive explanation of how a SOM is trained is provided, together with a formal explanation of the algorithm, and some of the more important parameters are discussed. Finally, an overview of different applications of SOMs in maritime problems is presented.
|Title of host publication||Information Fusion and Geographic Information Systems, Proceedings|
|Editors||VV Popovich, M Schrenk, C Claramunt, KV Korolenko|
|Place of Publication||Berlin|
|ISBN (Print)||1863-2246 978-3-642-00303-5|
|Publication status||Published - 1 Jan 2009|
|Name||Lecture Notes in Geoinformation and Cartography|