半导体光电, 2016, 37 (2): 275, 网络出版: 2016-05-11  

一种改进的CLIQUE高维子空间聚类算法

An Improved CLIQUE Subspace Clustering Algorithm for High Dimensional Data
作者单位
重庆理工大学 计算机科学与工程学院, 重庆 400054
摘要
分析了经典的CLIQUE聚类算法,阐述了该算法存在的局限性,针对该算法时间复杂度高和聚类精度低的问题,提出了一种改进的CLIQUE聚类算法;改进的算法不仅具有传统CLIQUE算法的优点,而且利用降低冗余维度和备份密集单元数据库D′的策略,大大降低了搜索成本和时间复杂度;且进一步用混合网格划分技术替代原有算法的固定网格划分技术,提高了聚类结果的精度,保留了密集单元的完整性。
Abstract
The classic clique clustering CLIQUE algorithm was analyzed, and the limitations of CLIQUE algorithm were explained. In view of the problems of high time complexity and low accuracy of clustering, an improved algorithm CLIQUE was further proposed. It not only has the advantages of traditional algorithms, but also it uses a value to reduce redundancy dimensions and backup intensive unit database, which greatly reduces the search cost and time complexity. The new algorithm also adopts the hybrid grid technology to replace the original algorithm of fixed grid technology, which both improves the accuracy of the clustering algorithm and maintains the integrity of the dense cells.

王东, 王理想, 范伟. 一种改进的CLIQUE高维子空间聚类算法[J]. 半导体光电, 2016, 37(2): 275. WANG Dong, WANG Lixiang, FAN Wei. An Improved CLIQUE Subspace Clustering Algorithm for High Dimensional Data[J]. Semiconductor Optoelectronics, 2016, 37(2): 275.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!