• DocumentCode
    151860
  • Title

    Towards a component-based software architecture for genetic algorithms

  • Author

    Garzon Rodriguez, Leidy ; Diosa, Henry Alberto ; Rojas-Galeano, Sergio

  • Author_Institution
    Univ. Distrital Francisco Jose de Caldas, Bogota, Colombia
  • fYear
    2014
  • fDate
    3-5 Sept. 2014
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    We are motivated on the idea of whether a component-based software architecture for evolutionary algorithms would be feasible and advantageous. We believe that depending on the evolutionary computation model, software implementing these algorithms can be robustly built assembling loosely-coupled computational blocks, likewise hardware systems that are built gluing together prefabricated electronic components. We set about to develop an initial architecture with a focus on the genetic algorithm. The paper describes the analysis and design principles used, the obtained architecture, the resulting component specification and closes with a discussion about the benefits of this approach, as well as initial steps towards its implementation in a user-friendly platform for component-based visual programming. The complete portfolio of software models is available at:http://arquisoft.udistrital.edu.co/portal/web/guest/proy-compAG.
  • Keywords
    formal specification; genetic algorithms; object-oriented programming; software architecture; visual programming; component specification; component-based software architecture; component-based visual programming; evolutionary algorithms; genetic algorithms; loosely-coupled computational blocks; Business; Computer architecture; Genetic algorithms; Sociology; Software; Statistics; Visualization; Architecture and Software Engineering; Component-based Development; Genetic Algorithms;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing Colombian Conference (9CCC), 2014 9th
  • Conference_Location
    Pereira
  • Type

    conf

  • DOI
    10.1109/ColumbianCC.2014.6955331
  • Filename
    6955331