摘要
利用Hadoop的并行式处理技术改进了遗传算法.通过将种群划分为多个子种群,并将每个子种群由一个单一的MapReduce任务来处理,实现了遗传算法的并行化.通过解决OneMax问题的一系列实验,验证了基于Hadoop的遗传算法提高了运行效率与正确性.
The genetic algorithm was improved based on the parallel process technology of Hadoop. The parallel genetic algorithm is implemented through divided the population into several sub-populations and each sub-population was processed by a single MapReduce.It is tested that the genetic algorithm based on Hadoop improved the efficiency and the accuracy through solving the OneMax problem.
出处
《内蒙古师范大学学报(自然科学汉文版)》
CAS
北大核心
2015年第1期62-65,共4页
Journal of Inner Mongolia Normal University(Natural Science Edition)
基金
内蒙古自然科学基金资助项目(2013MS0910)
关键词
遗传算法
Hadoop
MapReduce
Hadoop
MapReduce
GA