计算机科学 ›› 2014, Vol. 41 ›› Issue (8): 245-249.doi: 10.11896/j.issn.1002-137X.2014.08.052
吴明晖,张红喜,金苍宏,蔡文明
WU Ming-hui,ZHANG Hong-xi,JING Cang-hong and CAI Wen-ming
摘要： 传统网格聚类算法聚类质量低,而密度聚类算法时间复杂度高。针对两类算法各自的缺点,结合它们的聚类思想提出了一种新的聚类算法。该算法提出了边缘度密度距作为新的密度度量,并在此基础上逐步确定了类的定义和聚类过程的定义。算法前期通过网格划分操作统计记录了待聚类数据的初始信息,以供随后的k近邻统计使用。在寻找聚类中心点时使用了桶排序的策略,使得算法能快速地选出下一个聚类中心点。随后的聚类步骤是迭代搜索并检验当前类中未检验的k近邻是否满足密度可达性来完成聚类。理论分析和实验测试的结果表明,该算法不仅保持了较高的聚类精度,而且有接近线性的低时间复杂度。
| [1] Escudero L F,Garín M A,Pérez G,et al．Scenario Cluster Decomposition of the Lagrangian dual in two-stage stochastic mixed 0-1 optimization [J]．Computers & Operations Research,2013,1(40):362-377 [2] Wang W,Yang J,Muntz R．STING:A statistical information grid approach to spatial data mining [C]∥ Proc．of the 23rd Very Large Databases Conf．(VLDB 1997)．Athens,Greece．1997:186-195 [3] Rakai L,Farshidi A,Behjat L,et al．A New Length-Based Algebraic Multigrid Clustering Algorithm [J]．VLSI Design,2012,2012 [4] Demirtas E A．A Data Envelopment-Based Clustering Approach for Public Sugar Factories in Privatizing Process [J]．Mathematical Problems in Engineering,2011,2011 [5] Xu X-W,Ester M,Kriegel H P,et al．A distribution-based clustering algorithm for mining in large spatial databases [C]∥Proc．14th Internat．Conf．on Data Eng．(ICDE 98)．Orlando,FL,1998:324-331 [6] Zhong Y-F,Zhang L-P．A New Fuzzy Clustering algorithmBased on Clonal S election for Land Cover Classification [J]．Mathematical Problems in Engineering,2011,2011 [7] Elbatta M T,Bolbol R M,Ashour W M．A Vibration Method for Discovering Density Varied Clusters [J]． ISRN Artificial Intelligence,2012,2012 [8] Karypis G,Han E H,Kumar V．CHAMELEON:A hierarchical clustering algorithm using dynamic modeling [J]．IEEE Computer,1999,2(8):68-75 [9] 孙吉贵,刘杰,赵连宇．聚类算法研究[J]．软件学报,2008,1(19):48-64 [10] 赵慧,刘希玉,崔海青．网格聚类算法[J]．计算机技术与发展,2010,9(20) [11] Bhattacharya G,Ghosh K,Chowdhury A S．An affinity-based new local distance function and similarity measure for KNN algorithm [J]．Pattern Recognition Letters,2012,33(3):356-363 [12] Emre Celebi M,Kingravi H A ,Vela P A．A comparative study of efficient initialization methods for the Kmeans clustering algorithm [J]．Expert Systems with Applications,2013,40(1):200-210 [13] Tsai C,Chiu C．Developing a feature weight self-adjustment me-chanism for a Kmeans clustering algorithm [J]．Computational Statistics & Data Analysis,2008,10(52):4658-4672 [14] Mok P,Huang H,Ylkwok E．A robust adaptive clustering analysis method for automatic identification of clusters [J]．Pattern Recognition,2012,45(8):3017-3033 |
| No related articles found! |
|
||
首页