摘要
针对人工蜂群算法容易陷入局部最优值,收敛到最优解速度慢的缺点,通过使用固定步长和可变步长,定义采蜜蜂搜索食物源的公式,提出了一种改进的人工蜂群算法.对四个标准测试函数仿真表明该改进算法提高了算法的优化性能.为了改善数据挖掘中聚类算法效率,从人工蜂群算法评价函数入手,使用凝聚度函数、分散度函数,将改进的人工蜂群算法用于解决聚类问题,对三个数据集测试表明新算法在聚类准确率方面有一定提高.
The artificial bee colony algorithm is easy to fall into local optimal value, and it has the shortcoming of slow con- vergence to the optimal solution. By using the fixed step size and variable step length and defining the formula of bees＇ searching for food source, an improved artificial colony algorithm was proposed. Simulation on four standard test functions shows that this new algorithm improves the optimization performance of the algorithm. In order to improve the efficiency of clustering algorithm in data mining, starting from artificial bee colony algorithm evaluation function, this artificial bee colony algorithm can be used to solve the problem of clustering by using condensation degree function and dispersion function. The tests on three data sets show that this algorithm enhances the clustering accuracy.
出处
《宜宾学院学报》
2016年第6期41-45,共5页
Journal of Yibin University
关键词
数据挖掘
人工蜂群算法
聚类
data mining
artificial bee colony algorithm
clustering
