Learning with missing data using Genetic Programming   [MD] [GP]

by

Backer, G.

Literature search on Evolutionary ComputationBBase ©1999-2013, Rasmus K. Ursem
     Home · Search · Adv. search · Authors · Login · Add entries   Webmaster
Note to authors: Please submit your bibliography and contact information - online papers are more frequently cited.

Info: The 1st Online Workshop on Soft Computing (WSC1) (Conference proceedings), 1996
Keywords:genetic algorithms, genetic programming, Machine learning, Missing data, Strongly Typed Genetic Programming STGP
Abstract:
Learning with imprecise or missing data [MD] has been a major challenge for machine learning. [ML] A new approach using Strongly Typed Genetic Programming [ST] [GP] is proposed here, which uses extra computations based on other input data to approximate the missing values. It eliminates the need for pre-processing and makes use of correlations between the input data. The decision process itself and the handling of unknown data can be extracted from the resulting program for an analysis afterwards. Comparing it to an alternative approach on a simple example shows the usefulness of this approach.
Notes:
Adds "unknown" data type to STGP. demo on iris classification problem (see discussion on WSC1 pages) email WSC1 organisers wsc@bioele.nuee.nagoya-u.ac.jp
URL(s):(G)zipped postscript
HTML

Review item:

Mark as doublet (will be reviewed)

Print entry



BibTex:
@InProceedings{backer:1996:WSC,
  author =       "Gerriet Backer",
  title =        "Learning with missing data using Genetic Programming",
  booktitle =    "The 1st Online Workshop on Soft Computing (WSC1)",
  year =         "1996",
  address =      "http://www.bioele.nuee.nagoya-u.ac.jp/wsc1/",
  month =        "19--30 " # aug,
  organisation = "Research Group on ECOmp of the Society of Fuzzy Theory
                 and Systems (SOFT)",
  publisher =    "Nagoya University, Japan",
  keywords =     "genetic algorithms, genetic programming, Machine
                 learning, Missing data, Strongly Typed Genetic
                 Programming STGP",
  URL =          "http://www.psych.nat.tu-bs.de/psych/gb/wsc1/contents.htm",
  url_2 =        "http://www.bioele.nuee.nagoya-u.ac.jp/wsc1/papers/files/backer.ps",
  url_3 =        "http://www.bioele.nuee.nagoya-u.ac.jp/wsc1/papers/files/backer.ps.gz",
  abstract =     "Learning with imprecise or missing data has been a
                 major challenge for machine learning. A new approach
                 using Strongly Typed Genetic Programming is proposed
                 here, which uses extra computations based on other
                 input data to approximate the missing values. It
                 eliminates the need for pre-processing and makes use of
                 correlations between the input data. The decision
                 process itself and the handling of unknown data can be
                 extracted from the resulting program for an analysis
                 afterwards. Comparing it to an alternative approach on
                 a simple example shows the usefulness of this
                 approach.",
  size =         "5 pages",
  notes =        "Adds {"}unknown{"} data type to STGP. demo on iris
                 classification problem (see discussion on WSC1 pages)
                 email WSC1 organisers wsc@bioele.nuee.nagoya-u.ac.jp",
}