计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (24): 136-142.DOI: 10.3778/j.issn.1002-8331.1708-0276

• 模式识别与人工智能 • 上一篇    下一篇

自动确定聚类中心的数据竞争算法

许家楠,张桂珠   

  1. 江南大学 物联网工程学院,江苏 无锡 214122
  • 出版日期:2018-12-15 发布日期:2018-12-14

Automatically determine data competition algorithm for cluster center

XU Jianan, ZHANG Guizhu   

  1. School of Internet of Things Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Online:2018-12-15 Published:2018-12-14

摘要: 针对数据竞争算法采用欧式距离计算相似度、人为指定聚类簇数以及聚类中心无法准确自动确定等问题,提出了一种自动确定聚类中心的数据竞争聚类算法。引入了数据场的概念,使得计算出的势值更加符合数据集的真实分布;同时,结合数据点的势能与局部最小距离形成决策图完成聚类中心点的自动确定;根据近邻原则完成聚类。在人工以及真实数据集上的实验效果表明,提出的算法较原数据竞争算法具有更好的聚类性能。

关键词: 数据竞争, 数据场, 自动聚类, 密度不均匀

Abstract: Aiming at the similarity of Euclidean distance calculation, the number of clustering clusters and the clustering center can not be determined automatically and accurately, a data competition clustering algorithm is proposed to automatically determine the clustering center. Firstly, the concept of the data field is introduced so that the calculated potential value is more consistent with the true distribution of the data set. At the same time, the automatic determination of the clustering center is completed by combining the potential energy of the data point with the local minimum distance to form the decision graph. Principle to complete the cluster. The experimental results on the artificial and real data sets show that the proposed algorithm has better clustering performance than the original data competition algorithm.

Key words: data competition, data field, automatic clustering, density inhomogeneous