SOMKmeans两阶段聚类算法及其应用.pdfVIP

562010 16 327 þ ü SOM+ K-means 两阶段聚类算法及其应用 周 欢, 李广明, 张高煜 ( , 201209) :在众多聚类算法中, K-means 和自组织神经网络( SOM) 是较为经典的2 个在分 2 种算法优缺点的基础上, 提出基于SOM 的K-means 两阶段聚类算法, 该算法根据SOM 算法自动聚类的优点得到初步聚类数目和各类中心点, 以此 作为K-means 算法的初始输入进一步聚类, 从而得到精确的聚类信息最后, 应用该算法对某地区电信家庭客户数据进行 分 , 结果表明该算法有较好的聚类效果 :聚类; 自组织神经网络; K-means; 细分 :T N911-34 : A : 1004-373X( 2010) 16-0113-04 SOM+ K-means Two-phase Clustering Algorithm and Its Application ZHOU Huan, I Guang-m ing, ZHANG Gao-yu (School of Information M anagement, Shanghai Finance University, Shanghai 201209, China) Abstract: K-means and SOM network are tw o classical algorithms among many clustering ones. A new SOM-based K- means tw o-phase clustering algorithm is proposed based on the analysis of the advantages and shortcomings of the two algo- rithms. The quantity of the preliminary clustering and the central point of each cluster were acquired with K-means algorithm, by means of the auto-clustering advantages of SOM algorithm. Taking the results as the init ial input of the K-means algorithm to make the further clustering, the accurate clustering results are gained. The data of the telecom family customers in a dis- trict is analyzed w ith the algorithm. The results confirm that the algorithm is better than SOM netw ork and K-means algo- rithms when they are separately used. Keyw ords: clustering; SOM netw ork; K-means; partition , 0 , [ 1] , , , , , , , SOM ( / [ 9] , K-means ) ( , )