计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (14): 153-157.DOI: 10.3778/j.issn.1002-8331.1703-0228

• 模式识别与人工智能 • 上一篇    下一篇

基于密度峰值优化的模糊C均值聚类算法

刘沧生,许青林   

  1. 广东工业大学 计算机学院,广州 510006
  • 出版日期:2018-07-15 发布日期:2018-08-06

Fuzzy C-means clustering algorithm based on density peak value optimization

LIU Cangsheng, XU Qinglin   

  1. College of Computer Science, Guangdong University of Technology, Guangzhou 510006, China
  • Online:2018-07-15 Published:2018-08-06

摘要: 针对传统模糊C均值聚类算法和基于K-means++优化聚类中心的模糊C均值算法存在初始聚类中心敏感、聚类速度收敛慢、聚类算法需要人为给定聚类数目等缺陷,受密度峰值聚类算法(Clustering by Fast Search and Find of Density Peaks,CFSFDP)的启发,提出了基于密度峰值算法优化的模糊C均值聚类算法,自适应产生初始聚类中心,确定聚类数目,并优化算法收敛过程。实验结果表明,改进后的算法与传统模糊聚类C均值算法相比能够准确地得到簇的数目,性能有明显的提高,并加快算法的收敛速度,达到相对更好的聚类效果。

关键词: 模糊C均值聚类, 密度峰值, 密度聚类, 自适应

Abstract: Aiming at the traditional fuzzy C-means clustering algorithm and the fuzzy C-means algorithm based on K-means++ optimization clustering center, with the defects that initial clustering center sensitivity, clustering speed convergence is slow, the clustering algorithm needs to be given the number of artificial clustering, inspired by CFSFDP, a fuzzy C-means clustering algorithm based on density peak algorithm optimization is proposed. Adaptive clustering algorithm is generated to determine the number of clusters and to optimize the number of clusters. The clustering algorithm is based on the fast clustering algorithm(CFSFDP) convergence process. The experimental results show that the improved algorithm can accurately obtain the number of clusters and improve the performance compared with the traditional fuzzy clustering C - means algorithm, and accelerate the convergence speed of the algorithm to achieve a relatively better clustering effect.

Key words: fuzzy C-means clustering, density peak, density clustering, adaptive