From Bureau of Economic Geology, The
University of Texas at Austin (www.beg.utexas.edu).
For more information, please contact the author.
Self-organizing assemblages such as flocks of geese and schools of fish are examples of organizations whose members follow simple rules. With geese, the rule is to follow to the left or right of the goose in front and follow about a foot behind. While the rules for individuals are simple, the properties of assemblages are vastly different. Self-organizing Maps (SOM), discovered by T. Kohonen in the 1990's and brought to exploration geophysics by T. Taner and S. Treitel in the 2000's, are a type of computer learning machine which adapts to properties in the input data space. We illustrate SOM with simple examples from other fields and conclude with results where SOM has automatically identified seismic anomalies of geologic interest using multiple attributes of 3D seismic surveys.