计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (23): 141-142.
• 数据库、信号与信息处理 • 上一篇 下一篇
吴 佳，罗 可
收稿日期:
修回日期:
出版日期:
发布日期:
WU Jia，LUO Ke
Received:
Revised:
Online:
Published:
摘要： 针对FCM算法的缺点，提出了一种基于改进的FCM的增量式聚类方法。该算法首先对模糊C均值算法进行加权，并将权系数归一化，然后将改进的算法与增量式聚类算法结合。改进的方法既提高了FCM算法的性能，避免了FCM算法的缺陷，并能够实现增量式聚类，避免了大量的重复计算，并且不受孤立点的影响。实验表明该算法的有效性。
关键词: 聚类分析, 模糊C均值算法, 增量式聚类
Abstract: A new incremental clustering algorithm is proposed which is based on improved FCM for the shortcomings of FCM algorithm.The algorithm weights FCM which the weights will be normalized，then combines the improved algorithm and the incremental clustering algorithm.Improved method not only improves the performance and avoids the shortcomings of the FCM algorithm，but also be able to achieve incremental clustering，it avoids a lot of double counting and unaffected of outlier.Experiments show that the new algorithm is effective.
Key words: cluster analysis, Fuzzy C-Means（FCM） algorithm, incremental clustering
吴 佳，罗 可. 改进的模糊C均值的增量聚类算法[J]. 计算机工程与应用, 2011, 47(23): 141-142.
WU Jia，LUO Ke. Improved Fuzzy C-Means incremental clustering algorithm[J]. Computer Engineering and Applications, 2011, 47(23): 141-142.
0 / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://cea.ceaj.org/CN/
http://cea.ceaj.org/CN/Y2011/V47/I23/141
