摘要
借鉴蚁群算法的信息素机制,提出了一种基于信息素机制的离散粒子群算法。采用信息素机制的主要作用是使飞行在空间中的各个粒子不但要根据自己的信息来判断飞行方向,还可以根据其它粒子留下的信息进行方向判断。背包问题实验结果显示,该算法可以获得较优解。然后,将该算法应用到乳腺癌病人识别问题的特征选择上,结果显示,采用特征选择后的属性数据,所训练的网络可以获得较高的识别率。
A pheromone-based discrete particle swarm optimization algorithm was proposed borrowing the idea of pheromone refresh mechanism of ant colony algorithm.There was a main reason for using pheromone.The fly direction of particles in the space was changed according to not only the information of itself,but also the information other particles left. Knapsack problem was used to test the performance of the algorithm.Compared with the other algorithms,the results show that the proposed algorithm could acquire the better value.Then,the proposed algorithm is used to select characters in breast cancer recognition.The experimental results show that the recognition rate is obviously improved by using the attributes selected.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2008年第2期395-398,414,共5页
Journal of System Simulation
基金
国家自然科学基金资助项目(60675043)
关键词
离散粒子群
特征选择
信息素
背包问题
discrete particle swarm
character selection
pheromone
knapsack problem
