摘要
为改进传统K-means聚类算法中存在因随机选择初始质心而导致聚类结果不稳定且准确度低的缺点,提出基于改进量子旋转门人工鱼群算法的K-means聚类(IQAFSA)算法,通过动态更新量子旋转门的旋转角提高下一代更新方向准确度及更新速度。变异策略从传统的非门改为H门,既增加种群的多样性,又使全局搜索能力增强;最终使用所改进算法选取K-means的初始质心再进行聚类。通过UCI数据的测试以及在医学相关数据上的实验表明,提出的算法具有有效性,准确度较高且收敛速度较快。
To improve the instability and low accuracy of the clustering results caused by random selection of the initial centroid in traditional K-means clustering algorithm, this paper proposed a K-means clustering algorithm(IQAFSA) based on the improved quantum rotated gate artificial fish swarm algorithm. The IQAFSA improved the accuracy of the updating direction and the updating speed in next generation by dynamic updating the rotation angle of the quantum rotating gate. The mutation strategy changed from traditional Not-gate to H-gate, which not only increased the diversity of the population but also enhanced the global search ability. Finally, the improved algorithm was used to select the initial centroid of K-means for clustering. It was tested on UCI dataset and applied on medical related data sets. Experiment results show that the IQAFSA algorithm is effective, higher accuracy and faster convergence.
作者
白丽丽
宋初一
许丽艳
宋泽瑞
姜静清
Bai Lili;Song Chuyi;Xu Liyan;Song Zerui;Jiang Jingqing(College of Computer Science&Technology,Inner Mongolia Minzu University,Tongliao Inner Mongolia 028000,China;College of Mathematics&Physics,Inner Mongolia Minzu University,Tongliao Inner Mongolia 028000,China)
出处
《计算机应用研究》
CSCD
北大核心
2022年第3期797-801,806,共6页
Application Research of Computers
基金
国家自然科学基金资助项目(61662057,6216070371)。
关键词
聚类
K-MEANS
量子人工鱼群算法
量子旋转门
clustering
K-means
quantum artificial fish swarm algorithm
quantum rotating gate