摘要
由于布谷鸟算法的步长控制因子和发现概率在算法运行过程中保持固定,影响算法的整体寻优效率和寻优精度,为此提出一种自适应设置步长控制因子和发现概率的布谷鸟搜索算法,并利用它优化模糊聚类随机选取初始聚类中心影响聚类效果的缺陷。首先根据搜索阶段的不同自动调节两个参数,使全局和局部的搜索能力达到最平衡的状态,提高整体的搜索效率;然后用改进的布谷鸟搜索算法优化模糊聚类算法,使得算法达到更好的聚类效果。在对比实验中验证了改进后的自适应布谷鸟搜索算法在寻优速度和精度上效果更优。通过比较4种算法在UCI数据集上的聚类效果,验证了改进后的算法在聚类准确率和稳定性上都有所提升。
Considering the fact that both the step size control factor and the discovery probability of the cuckoo algorithm remains fixed in algorithm operation&which affects overall optimization efficiency and accuracy of the algorithm,a cuckoo search algorithm that adaptively sets the step size control factor and the discovery probability was proposed to optimize the defect that fuzzy clustering randomly selects the initial cluster center and affects the clustering effect.Firstly,having the two parameters automatically adjusted according to different search stages so that the global and local search capabilities can reach the most balanced state and improve the overall search efficiency;and then,having the improved cuckoo search algorithm used to optimize the fuzzy clustering algorithm so that the algorithm can achieve better clustering effect.The contrast experiment verified that the improved adaptive cuckoo search algorithm is more effective in searching speed and accuracy.Comparing the clustering effects of each algorithm on the UCI dataset verified that the improved algorithm has been improved in the clustering accuracy and stability.
作者
高妍妍
缪祥华
GAO Yan-yan;MIAO Xiang-hua(Faculty of Information Engineering and Automation;Yunnan Key Laboratory of Computer Technology Application&Kunming University of Science and Technology)
出处
《化工自动化及仪表》
CAS
2022年第6期725-731,共7页
Control and Instruments in Chemical Industry
关键词
模糊聚类
布谷鸟搜索
自适应优化控制
全局寻优
fuzzy clustering
cuckoo search
self-adaptive optimal control
global optimization
