Applied Chance Discovery and Genetic Engineering to Find the mprovement Of Genetic Algorithms

• Main Authors: Shung-Yen Kao, 高崇晏
• Other Authors: Wen-Shiu Lin
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/65114395076512977131

碩士 === 輔仁大學 === 資訊管理學系 === 94 === Genetic Algorithms (GAs) have several important features that predestine them to solve design problems. Their main disadvantage however is the excessively long run-time is needed to delivery satisfactory results for large instances of complex design problems and without incorporating with sufficient knowledge, it does have a significant impact on the efficiency of the search for an optimal solution. The main aims of this paper are(1)to demonstrate that the effective and efficient application of the GA concept to design problem solving requires substitution of the basic GAs natural evolutionary by applying chance discovery and genetic engineering,(2) to propose and discuss rough set theory for chance discovery and genetic engineering,(3)to show how to apply the improvement of genetic algorithms to solve De Jong set of test function. In this paper, we see whether rough set theory can be used to reveal good gene schema and use Genetic Engineering (GE) concept consists of perturbing the natural evolution of the basic GA to some degree by implementing some goal-oriented deterministic or semi-deterministic decisions. On the intellectual level, showing the connection between rough set theory and genetic engineering to improve GAs as related pieces of the innovation puzzle is scientifically and computationally interesting. The paper goes beyond mere conjecture and show how rough set theory can work together to find effective and efficient GAs.