Some Geographic Applications of Genetic Programming on the Cray T3D Supercomputer

Created by W.Langdon from gp-bibliography.bib Revision:1.4420

  author =       "I. Turton and S. Openshaw and G. Diplock",
  title =        "Some Geographic Applications of Genetic Programming on
                 the {Cray} T3D Supercomputer",
  booktitle =    "UK Parallel'96",
  year =         "1996",
  editor =       "Chris R. Jesshope and Alex V. Shafarenko",
  pages =        "135--150",
  address =      "University of Surrey",
  month =        "3-5 " # jul,
  organisation = "BCS Parallel Processing Specialist Group",
  publisher =    "Springer",
  keywords =     "genetic algorithms, genetic programming",
  ISBN =         "3-540-76068-7",
  isbn13 =       "978-3-540-76068-9",
  URL =          "",
  URL =          "",
  DOI =          "doi:10.1007/978-1-4471-1504-5_10",
  size =         "16 pages",
  abstract =     "The paper describes some geographical applications of
                 a parallel GP code which is run on a Cray T3D 512
                 processor supercomputer to create new types of well
                 performing mathematical models. A series of results are
                 described which allude to the potential power of the
                 method for which there are many practical applications
                 in spatial data rich environments where there are no
                 suitable existing models and no soundly based
                 theoretical framework on which to base them",
  notes =        "'Replacement of ephemeral constant by a parameter, the
                 value of which is optimised using an embedded non-linar
                 paramter estimation proceedure.' Spatial Interaction
                 Modelling, Seattle, USA Subglacial Water System Case
                 Study. Trapridge Glacier, Yukon.

                 From messages to
                 Tue, 03 Sep 1996 08:52:15 BST Thu, 05 Sep 1996 08:38:51
                 BST Basically we use an asyncronous steady state system
                 which implicitly gives preferential treatment to faster
                 evaluating models. Its written in F77 and MPI at the
                 moment but RSN I will be reworking it as F90 with some
                 more functionallity such as user configurable islands
                 and such like.

        Tue, 10 Sep 1996
                 08:32:17 BST I have produced a program that converts
                 s-experessions to infix equations that can be pasted
                 into maple (or free equivalents, I'd guess) which has a
                 simplify command. My code is at

Genetic Programming entries for Ian Turton Stanley Openshaw Gary Diplock