• DocumentCode
    387586
  • Title

    A framework of AIS based pattern classification and matching for engineering creative design

  • Author

    Xiao, Ren-bin ; Wang, Lei ; Liu, Yong

  • Author_Institution
    CAD Center, Huazhong Univ. of Sci. & Technol., Wuhan, China
  • Volume
    3
  • fYear
    2002
  • fDate
    2002
  • Firstpage
    1554
  • Abstract
    The artificial immune system (AIS) is a newly emerged biological computing technology that draws inspiration from vertebrate immune systems. From the computation point of view, AIS is powerful at learning and recognition; hence, a novel paradigm for machine learning and pattern recognition forms. Engineering creative design can be viewed as constraint satisfaction problems, which is NP-hard in essence, thereby it is generally hard for classical methods to solve. Since AIS has great potentials in solving hard and complex problems, it motivates the authors to explore the field of creative mechanism pattern synthesis, one of the most challenging fields in engineering creative design, with AIS. Specially, an engineering framework of AIS based pattern classification and matching is proposed, and the preliminary experiment results show the validity of the proposed method.
  • Keywords
    computational complexity; intelligent design assistants; learning (artificial intelligence); pattern classification; pattern clustering; pattern matching; NP-hard problems; artificial immune system; clustering; constraint satisfaction problems; curve pattern matching; engineering creative design; machine learning; pattern recognition; Artificial immune systems; Biology computing; Computers; Design engineering; Immune system; Machine learning; Pattern classification; Pattern matching; Pattern recognition; Power engineering and energy;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
  • Print_ISBN
    0-7803-7508-4
  • Type

    conf

  • DOI
    10.1109/ICMLC.2002.1167471
  • Filename
    1167471