| 基于公理化模糊子集的改进谱聚类算法 |
| |
| 引用本文： | 赵小强,刘晓丽.基于公理化模糊子集的改进谱聚类算法[J].电子与信息学报,2018,40(8):1904-1910. |
| |
| 作者姓名： | 赵小强 刘晓丽 |
| |
| 基金项目： | 国家自然科学基金(61763029)，甘肃省基础研究创新群体基金(1506RJIA031) |
| |
| 摘 要： | 谱聚类算法通常是采用高斯核作为相似性度量，并利用所有可用的特征来构建具有欧氏距离的相似度矩阵，数据集复杂度会影响其谱聚类性能，因此该文提出一种基于公理化模糊子集(AFS)的改进谱聚类算法。首先结合AFS算法，利用识别特征来衡量更合适的数据成对相似性，生成更强大的亲合矩阵；再有效地利用Nystr?m采样算法，计算采样点间以及采样点和剩余点间的相似度矩阵去降低计算的复杂度；最后通过在不同数据集以及图像分割上进行实验，证明了提出算法的有效性。
|
| 关 键 词： | 亲和矩阵 谱聚类 公理化模糊子集 Nystr?m采样算法 |
| 收稿时间： | 2017-09-25 |
An Improved Spectral Clustering Algorithm Based on Axiomatic Fuzzy Set |
| |
| Xiaoqiang ZHAO,Xiaoli LIU.An Improved Spectral Clustering Algorithm Based on Axiomatic Fuzzy Set[J].Journal of Electronics & Information Technology,2018,40(8):1904-1910. |
| |
| Authors: | Xiaoqiang ZHAO Xiaoli LIU |
| |
| Affiliation: | 1.College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China2.Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou 730050, China3.National Experimental Teaching Center of Electrical and Control Engineering, Lanzhou University of Technology, Lanzhou 730050, China |
| |
| Abstract: | Gaussian kernel is usually used as the similarity measure in spectral clustering algorithm, and all the available features are used to construct the similarity matrix with Euclidean distance. The complexity of the data set would affect its spectral clustering performance. Therefore, an improved spectral clustering algorithm based on Axiomatic Fuzzy Set (AFS) is proposed. Firstly, AFS algorithm is combined to measure the similarity of more suitable data by recognizing features, and the stronger affinity matrix is generated. Then Nystr?m sampling algorithm is used to calculate the similarity matrix between the sampling points and the sampling points and the remaining points to reduce the computational complexity. Finally, the experiment is carried out by using different data sets and image segmentations, the effectiveness of the proposed algorithm are proved. |
| |
| Keywords: | |
|
| 点击此处可从《电子与信息学报》浏览原始摘要信息 |
|
点击此处可从《电子与信息学报》下载全文 |
