| 一种改进的半监督谱聚类算法 |
| |
| 引用本文： | 王磊.一种改进的半监督谱聚类算法[J].商洛师范专科学校学报,2013(4):55-58,78. |
| |
| 作者姓名： | 王磊 |
| |
| 作者单位： | 商洛学院现代教育技术中心,陕西商洛726000 |
| |
| 摘 要： | 针对谱聚类算法稳定性较差的问题，提出了一种改进的半监督谱聚类算法。该算法依据图像的颜色、纹理和空间特征进行聚类，通过Bayes距离学习对相似度矩阵的内容进行修正；然后，使用半监督K—means聚类算法对调整后的特征向量进行聚类划分。仿真实验结果表明。较传统谱聚类而言该算法在准确率及稳定性上都有了显著提升。
|
| 关 键 词： | Bayes决策 谱聚类 稳定性 半监督 |
An Improved Semi-supervised Spectral Clustering Algorithm |
| |
| WANG Lei.An Improved Semi-supervised Spectral Clustering Algorithm[J].Journal of Shangluo Teachers College,2013(4):55-58,78. |
| |
| Authors: | WANG Lei |
| |
| Institution: | WANG Lei (Modem Education Technology Center, Shangluo Univeristy, Shangluo, Shaanxi 726000) |
| |
| Abstract: | An improved semi-supervised spectral clustering algorithm, as to its low stability of spectral clustering algorithm, is proposed. This algorithm is clustered according to color, texture and spatial characteristics of the image. It first adjusts the similarity matrix by distance learning methods based on Bayes decision; Then, the constrained K-means is used to cluster adjusted feature vectors. The simulation shows that the algorithm has traditional spectral clustering. significantly improved stability and accuracy than the |
| |
| Keywords: | Bayesian decision spectral clustering stability semi-supervised |
| 本文献已被 维普 等数据库收录！ |
|
