期刊文献+

基于质心的样本加权聚类算法

Algorithm of Sample's Weighting Clustering Based on Centroid
下载PDF
导出
摘要 针对传统的以k-means为代表的分割聚类算法认为所有的聚类样本对聚类中心的影响都是相同的这一观点,提出基于样本加权的聚类算法,并采用实际数据集验证算法的有效性.实验表明,该算法比传统的k-means聚类算法具有更高的精确度. The traditional partitional clustering algorithm represented by k-means considers all the clustering samples have the same impacts on clustering center.In view of this point,a clustering algorithm was proposed to deal with different samples' weight and was applied to the evaluation of teaching quality.To evaluate the proposed algorithm,some real and artificial dataset were used to verify the effectiveness of the algorithm.The findings show that the proposed algorithm has higher precision than the traditional k-means clustering algorithm.
出处 《成都大学学报(自然科学版)》 2011年第2期168-170,共3页 Journal of Chengdu University(Natural Science Edition)
基金 云南省教育厅科研基金(7C40843)资助项目
关键词 K-MEANS算法 聚类 样本加权 质心 k-means algorithm clustering sample's weighting centroid
  • 引文网络
  • 相关文献

参考文献5

  • 1Han J, Kamber M, Tung A K H. Spatial Clustering Methods inData Mining:A Survey[M] .New York:John Wiley,2001.
  • 2Jain A K, Murty M N, Flynnl P J. Data clustering: A review[J]. ACM Computing Surveys, 1999,31 (3) : 1145-1149.
  • 3Zhang T, Ramakrishnan R, Linvy M. Birch : an Efficient DataClustering Method for very large Databases [C]// Proceeding ofACM SIGMOD International Conference. New York: ACM Press,1996.
  • 4Ester M, Kriegel H, Sander J, et al. A Density-based Algorithmfor Discovering Clusters in Large Spatial Databases with Noise[C]//Proceed/ng of 2 nd International Conference on KnowledgeDiscovery and Data Mining. Portland, Oregon: ACM Press, 1996.
  • 5Wang W, Yang J, Myntz R. Sting: a Statistical Information GridApproach to Spatial Data Mining[C]//Proceedings of the 23 In-ternational Conference on Very Large Databases Athens. NewYork:ACMPress,1997.

相关主题

;
使用帮助 返回顶部