Interactive genetic algorithms with multi-population adaptive hierarchy and their application in fashion design

• Published in: Applied mathematics and computation 2007-02, Vol.185 (2), p.1098-1108
• Main Authors: Gong, Dun-Wei, Hao, Guo-Sheng, Zhou, Yong, Sun, Xiao-Yan
• Format: Article
• Language: English
• Subjects: Adaptation; Combinatorics; Combinatorics. Ordered structures; Designs and configurations; Exact sciences and technology; Fashion design; Genetic algorithms; Hierarchy; Interaction; Logic and foundations; Mathematical analysis; Mathematical logic, foundations, set theory; Mathematics; Multi-population; Numerical analysis; Numerical analysis. Scientific computation; Recursion theory; Sciences and techniques of general use
• ISSN: 0096-3003; 1873-5649
• DOI: 10.1016/j.amc.2006.07.043

Limitations of existing interactive genetic algorithms are analyzed and interactive genetic algorithms with multi-population adaptive hierarchy proposed. A model for interactive genetic algorithms with multi-population is established and a strategy for individuals’ migration is designed. Adaptive genetic operators are applied to interactive genetic algorithms with a single population, and when a condition for hierarchy is met, the algorithms will evolve in the subspace of the original search space. The algorithms’ implementation based on local networks is also presented. The algorithms proposed in this paper can maintain the diversity of populations as a whole, improve abilities in exploitation and exploration, prevent those good individuals from being eliminated and alleviate users’ fatigue. The application of the algorithms in fashion design validates that they are feasible and efficient.