基于萤火虫优化的加权K-means算法
针对传统K-means算法易受初始聚类中心和异常数据的影响等缺陷,利用萤火虫优化算法全局搜索能力强、收敛速度快的优势,对K-means算法的初始聚类中心进行优化,并通过引用一种加权的欧氏距离,减少异常数据等不确定因素带来的不良影响,提出了一种基于萤火虫优化的加权K-means算法。该算法在提升聚类性能的同时,有效增强了算法的收敛速度。在实验阶段,通过UCI数据集中的几组数据对该算法进行了聚类实验及有效性测试,实验结果充分表明了该算法的有效性及优越性。...
Saved in:
| Published in | 计算机应用研究 Vol. 35; no. 2; pp. 466 - 470 |
|---|---|
| Main Author | |
| Format | Journal Article |
| Language | Chinese |
| Published |
山东财经大学数学与数量经济学院,济南250014
2018
山东师范大学信息科学与工程学院,济南250014 山东警察学院公共基础部,济南250014%山东师范大学信息科学与工程学院,济南250014 山东省分布式计算机软件新技术重点实验室,济南250014%山东省分布式计算机软件新技术重点实验室,济南250014 |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Summary: | 针对传统K-means算法易受初始聚类中心和异常数据的影响等缺陷,利用萤火虫优化算法全局搜索能力强、收敛速度快的优势,对K-means算法的初始聚类中心进行优化,并通过引用一种加权的欧氏距离,减少异常数据等不确定因素带来的不良影响,提出了一种基于萤火虫优化的加权K-means算法。该算法在提升聚类性能的同时,有效增强了算法的收敛速度。在实验阶段,通过UCI数据集中的几组数据对该算法进行了聚类实验及有效性测试,实验结果充分表明了该算法的有效性及优越性。 |
|---|---|
| Bibliography: | 51-1196/TP weighted K-means ; clustering ; firefly algorithm The traditional K-means algorithm is sensitive to initial cluster and data noise, etc. To overcome these shortages, this paper used the firefly algorithm （FA） which had power ability of global search and quick convergence rate to optimize the initial clustering centers of traditional K-means algorithm. As the same time, it used a kind of weighted Euclidean distance to reduce the defects produced by noise data and other uncertainties. Thus it proposed a weighted K-means clustering algorithm based on FA, which could improve clustering performance as well as the convergence rate of the algorithm. Finally, this paper conducted clustering experiments and validity test on several groups of UCI data. The results show great efficiency and superio- rity of the proposed algorithm. Chen Xiaoxue1,2, Wei Yongqing2,3 , Ren Min1,4 , Meng Yuanyuan1,2（ 1. School of Information Science ＆ Engineering, Shandong Normal University, Jinan 250014, China; 2. Shandong Provincia |
| ISSN: | 1001-3695 |
| DOI: | 10.3969/j.issn.1001-3695.2018.02.031 |