| 一种改进的Fuzzy c—means聚类算法 |
| |
| 引用本文： | 胡钟山,丁震.一种改进的Fuzzy c—means聚类算法[J].南京理工大学学报(自然科学版),1997,21(4):337-340. |
| |
| 作者姓名： | 胡钟山 丁震 |
| |
| 作者单位： | 南京理工大学信息学院 |
| |
| 基金项目： | 国家教委优秀青年教师基金,国家教委回国人员基金,江苏省自然科学基金 |
| |
| 摘 要： | 该文提出了一种改进的ｆｕｚｚｙ ｃ－ｍｅａｎｓ算法（ＭＦＣＭ）。此算法是将传统算法（ＦＣＭ）直接对样本集聚类变为对特征集聚类，从而极大提高了ｆｕｚｚｙ ｃ－ｍｅａｎｓ的速度，证明了ＭＣＭ与ＦＭＣ在分类效果上的等价性，且ＭＦＣＭ较ＦＣＭ有较低的时间复杂性，讨论了ＭＦＣＭ与ＦＭＣ空间复杂性的关系。最后数值实验证实了结论。
|
| 关 键 词： | 模糊聚类 模式识别 聚类分析 MFCM |
A Modified Fuzzy c means Clustering Algorithm |
| |
| Hu ZhongshanDing ZhenYang JingyuTang ZhenmingWu Yongge.A Modified Fuzzy c means Clustering Algorithm[J].Journal of Nanjing University of Science and Technology(Nature Science),1997,21(4):337-340. |
| |
| Authors: | Hu ZhongshanDing ZhenYang JingyuTang ZhenmingWu Yongge |
| |
| Abstract: | In this paper a modified fuzzy c means algorithm (MFCM) is presented. MFCM uses feature set, instead of sample set, to cluster, and the computation time is greatly reduced. It is proved that the equality of FCM and MFCM in the clustering effect and the less time complexity in MFCM than that in FCM. Then it discussed the space complexity of MFCM and FCM. Finally, some experiment results are given to show the effectiveness of our algorithm. |
| |
| Keywords: | fuzzy clustering pattern recognition cluster analysis image processing fuzzy c means algorithm |
| 本文献已被 CNKI 维普 等数据库收录！ |
