Discussion of Offspring Generation Method for Interactive Genetic Algorithms with Consideration of Multimodal Preference

  • Fuyuko Ito
  • Tomoyuki Hiroyasu
  • Mitsunori Miki
  • Hisatake Yokouchi
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5361)


The interactive genetic algorithm(iGA) is a method to obtain and predict a user’s preference based on subjective evaluation of users, and it has been applied to many unimodal problems, such as designing clothes or fitting of hearing aids. On the other hand, we are interested in applying iGA to user’s preferences, which can be described as a multimodal problem with equivalent fitness values at the peaks. For example, when iGA is applied to product recommendation on shopping sites, users have several types of preference trends at the same time in product selection. Hence, reflecting all the trends in product presentation leads to increased sales and consumer satisfaction. In this paper, we propose a new offspring generation method that enables efficient search even with multimodal user preferences by introducing clustering of selected individuals and generating offspring from each cluster. Furthermore, we perform a subjective experiment using an experimental iGA system for product recommendation to verify the efficiency of the proposed method. The results confirms that the proposed method enables offspring generation with consideration of multimodal preferences, and there is no negative influence on the performance of preference prediction by iGA.


Design Variable Design Space Product Recommendation Generation Range Multimodal Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Takagi, H.: Interactive evolutionary computation: Fusion of the capabilities of ec optimization and human evaluation. Proceedings of the IEEE 89(9), 1275–1296 (2001)CrossRefGoogle Scholar
  2. 2.
    Goldberg, D.E.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley Professional, Reading (1989)zbMATHGoogle Scholar
  3. 3.
    Llorà, X., Sastry, K., Goldberg, D.E., Gupta, A., Lakshmi, L.: Combating user fatigue in igas: partial ordering, support vector machines, and synthetic fitness. In: Proceedings of Genetic Evolutionary Computation Conference (GECCO), pp. 1363–1370 (2005)Google Scholar
  4. 4.
    Konstan, J.A., Miller, B.N., Maltz, D., Herlocker, J.L., Gordon, L.R., Riedl, J.: Grouplens: applying collaborative filtering to usenet news. Commun. ACM 40(3), 77–87 (1997)CrossRefGoogle Scholar
  5. 5.
    Sarwar, B., Karypis, G., Konstan, J., Reidl, J.: Item-based collaborative filtering recommendation algorithms. In: WWW 2001: Proceedings of the 10th international conference on World Wide Web, pp. 285–295. ACM, New York (2001)Google Scholar
  6. 6.
    Aoki, K., Takagi, H.: Interactive ga-based design support system for lighting design in 3-d computer graphics. The transactions of the Institute of Electronics, Information and Communication Engineers 81(7), 1601–1608 (1998)Google Scholar
  7. 7.
    Hocaoglu, C., Sanderson, A.C.: Multimodal function optimization using minimal representation size clustering and its application to planning multipaths. Evolutionary Computation 5(1), 81–104 (1997)CrossRefGoogle Scholar
  8. 8.
    Handl, J., Knowles, J.: Improving the scalability of multiobjective clustering. In: Proceedings of the Congress on Evolutionary Computation (CEC 2005), vol. 3, pp. 2372–2379 (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Fuyuko Ito
    • 1
  • Tomoyuki Hiroyasu
    • 2
  • Mitsunori Miki
    • 3
  • Hisatake Yokouchi
    • 2
  1. 1.Graduate School of EngineeringDoshisha UniversityKyotoJapan
  2. 2.Department of Life and Medical SciencesDoshisha UniversityJapan
  3. 3.Department of Science and EngineeringDoshisha UniversityJapan

Personalised recommendations