| B. Nassersharif, D. Ence, and M. Au. Visualisation of evolution of genetic algorithms. In The Proceedings of the World Congress on Neural Networks, volume 1, pages 560-565, 1994. |
....connections between the cities change, and could be used to show the current best fit phenotype. Finally, the fifth category deals with visualising the GA s sampling of the search space. Because a non ambiguous mapping is not possible beyond 2d problems, the early work concentrated on this [Nassersharif et al. 1994]. Other studies used Principal Component Analysis [Harvey and Thompson, 1996] see figure 1.7) Sammon Mapping [Dybowski et al. 1996] or other projection tech niques. They project the genotypic space onto a 2d or 3d space, and draw a scatterplot of all individuals, or those in the current ....
B. Nassersharif, D. Ence, and M. Au. Visualisation of evolution of genetic algorithms. In The Proceedings of the World Congress on Neural Networks, volume 1, pages 560-565, 1994.
.... with modern human computer interaction technology to facilitate the human understanding and effective use of computer software [19] The application of SV techniques to facilitate the design and application of EAs has been receiving growing attention during the last few years; 2] 12] 20] [16], 7] 9] 21] and [5] A widely recognized framework for understanding human memory, known as the Levels of processing framework, was proposed by Craik and Lockhart in 1972 [6] They suggested that the level or depth of processing of a stimulus has a substantial effect on its memorability, ....
B. Nassersharif, D. Ence, and M. Au. Visualisation of evolution of genetic algorithms. In , volume 1, pages 1--560--1-- 565, San Diego, CA., USA, 1994.
....known approach to visualizing the chromosome data is the use of a dataspace metaphor in which each chromosome is represented as a point in two or three dimensional space mapped from its position in a higherdimensional allele space. This creates a simple visual image of a GA s exploratory search. Nassersharif et al. (1994) proposed this method for two dimensional problems in which the GA s chromosomes could be displayed directly in a three dimensional scatterplot. The two problem dimensions of each chromosome were mapped onto the x and y axes, with the corresponding fitness rating being mapped onto the z axis. ....
....or disc plots, it can provide contour diagrams. However, a problem with this method is that it can involve considerable computational time thereby slowing down the rate with which the progress of a GA is displayed or analysed. 6. 0 Conclusion Our approach goes beyond those of Collins (1993) and Nassersharif et al. (1994). It makes the visualization of a GA s chromosomes a possibility, illustrating the number of solutions being considered, their similarity and fitness ratings. In particular, we demonstrated in Section 4.2 how the technique can identify the occurrence of alternative solutions. Because of the ....
Nassersharif B, Ence D, Au M (1994). Visualisation of evolution of genetic algorithms. In Proceedings of the World Congress on Neural Networks. Hillside, NJ: Lawrence Erlbaum Associates. pp I-560-I-565.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC