Creative design by chance based interactive evolutionary computation

  • Authors:
  • Chao-Fu Hong;Hsiao-Fang Yang;Mu-Hua Lin

  • Affiliations:
  • Associate Proffessor, Department of Information Management, Altheia University, Taipei, Taiwan;Graduate School of Management Sciences, Altheia University, Taipei, Taiwan;Graduate School of Management Sciences, Altheia University, Taipei, Taiwan

  • Venue:
  • ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part III
  • Year:
  • 2005

Quantified Score

Hi-index 0.01

Visualization

Abstract

Kotler and Trias De Bes (2003) at Lateral Marketing defined the creativity: each cluster had its own concepts, when a new need was generated and the designer could not find a solution from his own clusters, therefore he had a gap to overcome. This gap was as the original of creativity. If he wants to solve this problem, chose a new important concept for beginning was the only way he could do. This phenomenon was called the laterally transmitting. Then according to the designer's subjective to choose the concept and connected other cluster to generate or enter a new cluster. This kind designing process could generate a creative product. But it also brought a new problem, there had many concepts in conceptual space, how to decide an effectiveness concept and extents it to create a good product. Here we combined Watt's (2003) Small World model and Ohsawa and McBurney (2003) Chance Discovery Model to decide the creative probability and decreased the searching path length. Finally, we integrated the choosing mechanism and recombination mechanism into our chance based IEC (CBIEC) model. And we applied on the cell phone design. After the experiment we analyzed the interactive data found the choosing mechanism could bring the effective creativity and the recombination mechanism could quickly search as we expected the short-cut effect. Beside these results we also directly investigated the subjective of designer found our CBIEC model also better than the IGA (interactive genetic algorithms, Caldwell and Johnston, 1991).