 |
    |
New Usage of Sammon's Mapping for Genetic Visualization
Yong-Hyuk Kim and Byung-Ro Moon
School of Computer Science & Engineering,
Seoul National University
Shillim-dong, Kwanak-gu,
Seoul, 151-742 Korea
{yhdfly, moon}@soar.snu.ac.kr
Abstract. It is a hard problem to understand the fitness landscape of a problem as well as the evolution of genetic algorithms. For the purpose, we adopt Sammon's mapping for the investigation. We demonstrate its usefulness by applying it to the graph partitioning problem which is a well-known NP-hard problem. Also, through the investigation of schema traces, we explain the genetic process and the reordering effect in the genetic algorithm.
LNCS 2723, p. 1136 ff.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2003