计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (35): 51-53.

• 研究、探讨 • 上一篇    下一篇

面向自动组卷问题的改进类电磁算法

杨世达,金 敏,梅 磊   

  1. 武汉理工大学 计算机科学与技术学院,武汉 430070
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-12-11 发布日期:2011-12-11

Improved electromagnetism-like mechanism algorithm for automatic test paper problem

YANG Shida,JIN Min,MEI Lei   

  1. School of Computer Science and Technology,Wuhan University of Technology,Wuhan 430070,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-12-11 Published:2011-12-11

摘要: 类电磁算法(EM)中局部搜索是按一定步长进行线性搜索,在这个范围内寻找个体在某一维上的最优值。由于步长的限定,求得的该维上最优值可能远离实际的最优值。采用遗传算法(GA)中选择因子和交叉因子可以很好地解决这一问题。在组卷系统中,通过基于遗传算法改进的类电磁算法(Based Genetic Electromagnetism-like Mechanism Algorithm,GEM)与GA算法以及采用线性局部搜索的EM算法实验的比较,证明该算法有更高的组卷效率。

关键词: 类电磁算法, 全局优化, 遗传算法, 自动组卷, 局部搜索

Abstract: Local search of Electromagnetism-like Mechanism algorithm(EM) is linear searched by a certain step length,which finds the optimal value in a particular dimension in this individual.As the limit of step length,the optimal value may be far from the actual value of a particular dimension.To solve this problem,the select and crossover factor of Genetic Algorithm(GA) is the good way.In the automatic test paper system,compared with the Based Genetic Electromagnetism-like Mechanism Algorithm(GEM),GA and the EM of linear search in experiments show that the GEM has higher efficiency test paper.

Key words: electromagnetism-like mechanism algorithm, global optimization, genetic algorithm, automatic test paper, local search