Using software visualization technology to help evolutionary algorithm users validate their solutions.
In: Seventh International Conference on Genetic Algorithms (ICGA'97), 19 - 23 July 1997, East Lansing, MI, USA.
Evolutionary Computation (EC) offers a variety of robust search techniques, each of which operate by means of an evolutionary-based search algorithm. In order to assess the quality of any solutions found, the user must understand the search path taken to discover them. Software Visualization is recommended as a useful technique for aiding such understanding. This paper gives an introduction to Software Visualization, reviews five existing techniques for presenting EC output data and presents a new technique suitable for visualizing EC search space. To conclude a summary table of all six visualization techniques, including comments and criticisms of each is given and the selection of appropriate visualization techniques is discussed. By presenting EC output data in a format that is easy to understand software visualization makes EC validation much simpler for the user.
Actions (login may be required)