Method Trees: Building Blocks for Self-Organizable Representations of Value Series

Created by W.Langdon from gp-bibliography.bib Revision:1.4221

  author =       "Ingo Mierswa and Katharina Morik",
  title =        "Method Trees: Building Blocks for Self-Organizable
                 Representations of Value Series",
  booktitle =    "Genetic and Evolutionary Computation Conference
                 {(GECCO2005)} workshop program",
  year =         "2005",
  month =        "25-29 " # jun,
  editor =       "Franz Rothlauf and Misty Blowers and 
                 J{\"u}rgen Branke and Stefano Cagnoni and Ivan I. Garibay and 
                 Ozlem Garibay and J{\"o}rn Grahl and Gregory Hornby and 
                 Edwin D. {de Jong} and Tim Kovacs and Sanjeev Kumar and 
                 Claudio F. Lima and Xavier Llor{\`a} and 
                 Fernando Lobo and Laurence D. Merkle and Julian Miller and 
                 Jason H. Moore and Michael O'Neill and Martin Pelikan and 
                 Terry P. Riopka and Marylyn D. Ritchie and Kumara Sastry and 
                 Stephen L. Smith and Hal Stringer and 
                 Keiki Takadama and Marc Toussaint and Stephen C. Upton and 
                 Alden H. Wright",
  publisher =    "ACM Press",
  address =      "Washington, D.C., USA",
  keywords =     "genetic algorithms, genetic programming",
  pages =        "293--300",
  URL =          "",
  abstract =     "We introduce a framework for automatic feature
                 extraction from very large series. The extracted
                 features build a new representation which is better
                 suitable for a given learning task. The development of
                 appropriate feature extraction methods is a tedious
                 effort, particularly because every new classification
                 task requires tailoring the feature set anew.
                 Therefore, the simple building blocks defined in our
                 framework can be combined to complex feature extraction
                 methods. We employ a genetic programming approach
                 guided by the performance of the learning classifier
                 using the new representation. Our approach to evolve
                 representations from series data requires a balance
                 between the completeness of the methods on one side and
                 the tractability of searching for appropriate methods
                 on the other side. Some theoretical considerations
                 illustrate the trade-off. After the feature extraction,
                 a second process learns a classifier from the
                 transformed data. The practical use of the methods is
                 shown by two types of experiments in the domain of
                 music data classification: classification of genres and
                 classification according to user preferences.",
  notes =        "Distributed on CD-ROM at GECCO-2005. ACM

Genetic Programming entries for Ingo Mierswa Katharina Morik