Paper:
Views over last 60 days: 321
A Study on Self-Organization of Heuristics Using Analogy
Kenichi Matsuura and Yukinori Kakazu
Faculty of Engineering, Hokkaido University, Nishi 8, Kita 13-Jyo, Kita-ku, Sapporo 060, Japan
Received:January 31, 1996Accepted:February 25, 1996Published:August 20, 1996
Keywords:Self-organization, Heuristics, Analogy, Cluster analysis
Abstract
To solve a given problem, the strategy which is suitable for solving its problem is used in general. Those strategies for problem solving can be treated as mapping from the problem into a solution. Besides, there will exist some kind of analogical relationship among these strategies. Thus, we can apply the method, that takes analogy of strategies with the problem which have already been solved and which is similar to the unsolved problem, to search solution for unsolved target problem. In this paper, we attempt to perform problem solving using such analogical method. To this end, first, heuristics are organized by generalizing a set of strategies for problem solving which are similar to each other. Second, those heuristics are applied to the unsolved target problem to search a solution. We have performed some computer experiments to examine the role of heuristics and utility of analogy.
Cite this article as:K. Matsuura and Y. Kakazu, “A Study on Self-Organization of Heuristics Using Analogy,” J. Robot. Mechatron., Vol.8 No.4, pp. 351-355, 1996.Data files:
