谱聚类中的相似度矩阵研究 |
| |
引用本文: | 刘志伟.谱聚类中的相似度矩阵研究[J].电脑与微电子技术,2010(11):67-69. |
| |
作者姓名: | 刘志伟 |
| |
作者单位: | 温州大学物理与电子信息工学院,温州325035 |
| |
摘 要: | 以谱聚类算法为理论基础,对图的分割问题构造合适的相似度矩阵。构造四种相似度矩阵,并进行实验分析。实验结果表明,好的相似度矩阵对图的分割是非常有效的。
|
关 键 词: | 谱聚类 最短距离 边聚类系数 聚类 模式识别 |
Research on Similarity Matrix of Spectral Clustering |
| |
Authors: | LIU Zhi-wei |
| |
Affiliation: | LIU Zhi-wei(College of Physics & Electronic Information Engineering,Wenzhou University,Wenzhou 325035) |
| |
Abstract: | Taking spectral clustering algorithm as rationale,constructs suitable similarity degree matrix forgraph division problem.Constructs four kinds similarity degree matrix,and carries out the experiment analysis.It is indicated that good similarity degree matrix for the graph division is effective. |
| |
Keywords: | Spectral Clustering Shortest Distance Edge Clustering Coefficient Clustering Pattern Recognition |
本文献已被 维普 等数据库收录! |
|