Feature Construction and Dimension Reduction Using Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "Kourosh Neshatian and Mengjie Zhang and 
                 Mark Johnston",
  title =        "Feature Construction and Dimension Reduction Using
                 Genetic Programming",
  year =         "2007",
  editor =       "Mehmet A. Orgun and John Thornton",
  booktitle =    "Proceedings of the 20th Australian Joint Conference on
                 Artificial Intelligence",
  publisher =    "Springer",
  series =       "Lecture Notes in Computer Science",
  volume =       "4830",
  address =      "Gold Coast, Australia",
  month =        dec # " 2-6",
  pages =        "160--170",
  bibsource =    "DBLP, http://dblp.uni-trier.de",
  keywords =     "genetic algorithms, genetic programming",
  isbn13 =       "978-3-540-76926-2",
  DOI =          "doi:10.1007/978-3-540-76928-6_18",
  size =         "11 pages",
  abstract =     "This paper describes a new approach to the use of
                 genetic programming (GP) for feature construction in
                 classification problems. Rather than wrapping a
                 particular classifier for single feature construction
                 as in most of the existing methods, this approach uses
                 GP to construct multiple (high-level) features from the
                 original features. These constructed features are then
                 used by decision trees for classification. As feature
                 construction is independent of classification, the
                 fitness function is designed based on the class
                 dispersion and entropy. This approach is examined and
                 compared with the standard decision tree method, using
                 the original features, and using a combination of the
                 original features and constructed features, on 12
                 benchmark classification problems. The results show
                 that the new approach outperforms the standard way of
                 using decision trees on these problems in terms of the
                 classification performance, dimension reduction and the
                 learned decision tree size.",

Genetic Programming entries for Kourosh Neshatian Mengjie Zhang Mark Johnston