首页> 外文会议>International Conference on Natural Computation;ICNC '09 >A Coevolutionary Approach to Substructure Discovery Based on Individual Cooperation
【24h】

A Coevolutionary Approach to Substructure Discovery Based on Individual Cooperation

机译:基于个体合作的协同进化子结构发现方法

获取原文

摘要

A cooperative coevolutionary EA based algorithm is developed to discover potentially useful substructures from graphical databases. Unlike the usual coevolutionary algorithms which are based on the divide-and-conquer strategy with different populations representing different subtasks, the cooperation in our algorithm is at individual-level and implemented by a new genetic operator, the individual cooperation operator. The operator, during the searching process, enables different individuals to search the same substructure in a cooperative way and hence handles the problem of losing instances, which is very common and vital to the algorithm performance. In addition, an approximate graph matching algorithm is also proposed to make the operator more efficient. Experimental results show that the new operator successfully enhances the searching capability of the algorithm and improves the qualities of solutions.
机译:开发了一种基于协同协同进化EA的算法,以从图形数据库中发现潜在有用的子结构。与通常的基于不同群体代表不同子任务的“分而治之”策略的协同进化算法不同,我们算法中的协作是在个体级别上进行的,并且是由新的遗传算子(个体协作算子)实现的。在搜索过程中,运算符使不同的个体能够以协作的方式搜索相同的子结构,从而解决了丢失实例的问题,这对于算法性能而言是非常普遍且至关重要的。此外,还提出了一种近似图匹配算法,以使操作员更高效。实验结果表明,该新算子成功地提高了算法的搜索能力,提高了解的质量。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号