Advertisement

Information Space Optimization with Real-Coded Genetic Algorithm for Inductive Learning

  • Ryohei Orihara
  • Tomoko Murakami
  • Naomichi Sueda
  • Shigeaki Sakurai
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 14)

Abstract

New feature construction methods are presented. The methods are based on the idea that a smooth feature space facilitates inductive learning thus it is desirable for data mining The methods, Category-guided Adaptive Modeling (CAM) and Smoothness-driven Adaptive Modeling (SAM), are originally developed to model human perception of still images, where an image is perceived in a space of index colors. CAM is tested for a classification problem and SAM is tested for a Kansei scale value (the amount of the impression) prediction problem. Both algorithms have been proved to be useful as preprocess steps for inductive learning through the experiments. We also evaluate SAM using datasets from the UCI repository and the result has been promising.

Keywords

Linear Discriminant Analysis Index Color Transformation Function Image Perception Inductive Learning 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Rendell, L. (1990): Feature construction for concept learning. In Benjamin, D., ed.: Change of representation and inductive bias. Kluwer Academic, 327–353CrossRefGoogle Scholar
  2. 2.
    Murakami, T., et al. (2000): Friendly information retrieval through adaptive restructuring of information space. In: Proc. of AIE/IEA 2000, 639–644Google Scholar
  3. 3.
    Murakami, T., Orihara, R. (2000): Friendly information retrieval through adaptive restructuring of information space. New Generation Computing 18, 137–46Google Scholar
  4. 4.
    Blake, C., Merz, C.: UCI repository of machine learning databases. http://www.ics.uci.eduk-mlearn/MLRepository.html (1998)Google Scholar
  5. 5.
    Morohara, Y., et al. (1995): Automatic picking of index colors in textile pictures for designers. Trans. Inf. Process. Soc. Jpn. 36, 329–337, in Japanese.Google Scholar
  6. 6.
    Ikeda, M. (1980): Foundation of Color Engineering. Asakura Shoten, in Japanese.Google Scholar
  7. 7.
    Yamazaki, H., Kondo, K. (1998): A method of changing a color scheme with kansei scales. In: Proc. of 8th International Conference on Engineering Computer Graphics and Descriptive Geometry, 210–214Google Scholar
  8. 8.
    Muggleton, S. (1987): DUCE, an Oracle based Approach to Constructive Induction. In: Proc. of 10th IJCAI, 287–292Google Scholar
  9. 9.
    Higuchi, T., et al. (2001): Simplex crossover for real-coded genetic algorithms. Transactions of the JSAI 16, 147–155 in Japanese.MathSciNetGoogle Scholar
  10. 10.
    Torkkola, K., Campbell, W. (2000): Mutual information in learning feature transformations. In: Proc. of ICML 2000Google Scholar
  11. 11.
    Quinlan, J. (1993): C4. 5: Programs for Machine Learning. Morgan KaufmannGoogle Scholar
  12. 12.
    Kira, K., Rendell, L. (1992): The feature selection problem: Traditional method and a new algorithm. In: Proc. of AAAI’92, 129–134Google Scholar
  13. 13.
    Toshiba Corporation (1997): Data Mining Tool KINOsuite-PR. http://www2.toshiba.co.jp/datamining/index.htmGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Ryohei Orihara
    • 1
  • Tomoko Murakami
    • 1
  • Naomichi Sueda
    • 1
  • Shigeaki Sakurai
    • 1
  1. 1.RWCP Information-Base Functions Toshiba LaboratoryCorporate R&D Center, Toshiba Corp.Saiwai-ku, KawasakiJapan

Personalised recommendations