基于改进麻雀搜索算法的K-means聚类
K-means Clustering Based on Improved Sparrow Search
翁嘉诚 1周晓杰 1叶蓓蕾 1王建宏1
作者信息
- 1. 南通大学理学院,江苏 南通 226019
- 折叠
摘要
针对传统K-means聚类算法依赖初始解、易陷入局部最优等问题,提出了一种基于改进麻雀搜索算法的K-means聚类.首先,将分数阶微积分引入麻雀算法,用 自适应分数阶阶次对麻雀进行位置更新,提高算法的收敛精度;其次,对麻雀种群进行精英反向学习,增强种群多样性,扩大搜索区域范围;然后,对麻雀的位置进行自适应t分布变异,避免算法陷入局部最优;最后,将提出的改进算法在8个基准测试函数中进行性能验证,并应用于K-means聚类.对UCI数据集进行聚类仿真实验,结果表明基于改进麻雀搜索算法的K-means聚类可有效提高聚类质量和算法稳定性.
Abstract
Aiming at the problems that traditional K-means clustering algorithm relies on the initial solution and is easy to fall into local optima,a K-means clustering based on the improved sparrow search algorithm is proposed.First,introduce the fractional calculus into the sparrow algorithm,and update the position of the sparrow with an adaptive fractional order to improve the convergence accuracy of the algorithm;Secondly,perform elite reverse learning on the sparrow population to enhance the diversity of the population and expand the scope of the search area;Then,adaptive t distribution mutation is performed on the position of the sparrow to avoid the algorithm from falling into the local optimum;Finally,the performance of the proposed improved algorithm is verified in 8 benchmark functions and applied to K-means clustering.The clustering simulation experiment on the UCI data set shows that the K-means clustering based on the improved sparrow search algorithm can effectively improve the clustering quality and algorithm stability.
关键词
麻雀算法/自适应分数阶/精英反向学习/自适应t分布/K-means聚类Key words
sparrow algorithm/adaptive fractional order/elite reverse learning/adaptive t distribution/K-means clustering引用本文复制引用
基金项目
全国统计科学研究项目(2020LY020)
南通市科技计划项目(MS12021058)
出版年
2024
NETL
NSTL
万方数据