通信学报 ›› 2012, Vol. 33 ›› Issue (Z2): 290-293.doi: 10.3969/j.issn.1000-436x.2012.z2.042

• 学术论文 • 上一篇    下一篇

基于信息熵的蚁群聚类DBSCAN改进算法

张拥华,杜飞明,吴代文   

  1. 湖南工业职业技术学院 经济管理系,湖南 长沙 410208
  • 出版日期:2012-11-25 发布日期:2017-08-03
  • 基金资助:
    湖南省教育厅科研基金资助项目

Improved density-dased clustering algorithm based on information entropy and ant colony optimization abstract

Yong-hua ZHANG,Fei-ming DU,Dai-wen WU   

  1. Department of Economic Management,Hunan Industry Polytechnic,Changsha 410208,China
  • Online:2012-11-25 Published:2017-08-03
  • Supported by:
    The Scientific Research Foundation of Education Department of Hunan Province

摘要:

针对DBSCAN算法对数据分布不均匀和大规模数据处理问题上的不足,提出了一种新的整合算法,算法使用信息熵和蚁群聚类技术对聚类数据集进行代表性子集选择,在子集基础上进行DBSCAN聚类,实验证明这一算法能显著降低I/O耗费和内存需求,有效地解决含有分类属性的高维大规模数据集的聚类问题。

关键词: 信息熵, 聚类, DBSCAN, 蚁群算法

Abstract:

An integration of clustering algorithm was proposed for the shortage of the DBSCAN algorithm in inhomogeneous distribution and large-scale data processing.The algorithm extracted representative data from the original data set using information entropy and ant colony clustering technology,and did DBSCAN clustering based on the representative data subset.The experiment show that this algorithm is effective to reduce the I/O-consuming and memory requirements,and resolve the cluster problem of large-scale data sets containing categories property.

Key words: information entropy, clustering, DBSCAN, ant colony optimization

No Suggested Reading articles found!