| 一种基于马氏距离的半监督模糊聚类方法及其应用 |
| |
| 引用本文： | 施伟民,杨昔阳,李志伟.一种基于马氏距离的半监督模糊聚类方法及其应用[J].厦门大学学报(自然科学版),2012,51(3):311-315. |
| |
| 作者姓名： | 施伟民 杨昔阳 李志伟 |
| |
| 作者单位： | 泉州师范学院数学与计算机科学学院,福建泉州,362000 |
| |
| 基金项目： | 福建省教育厅A类科技项目，陕西省科技厅重点国际科技合作项目，福建省高校服务海西建设重点项目，泉州市科技计划项目 |
| |
| 摘 要： | 定义了一种新的基于马氏距离的半监督模糊聚类算法,并推导出它的迭代公式.该算法能够提高聚类算法的运行效率.在人工数据集和真实数据集上的实验结果验证了这种方法的有效性.提取了黄瓜叶片7个色调特征,Fisher降维之后进行半监督聚类分析.对于已标识类别属性的叶片,聚类结果与已知属性的一致率达100％,而对于未标识数据,一致率也达到96％以上.
|
| 关 键 词： | 马氏距离 半监督 模糊聚类 |
A Novel Semi Supervised Fuzzy Clustering Algorithm Based on Mahalanobis Distance |
| |
| SHI Wei-min
,
YANG Xi-yang
,
LI Zhi-wei.A Novel Semi Supervised Fuzzy Clustering Algorithm Based on Mahalanobis Distance[J].Journal of Xiamen University(Natural Science),2012,51(3):311-315. |
| |
| Authors: | SHI Wei-min YANG Xi-yang LI Zhi-wei |
| |
| Institution: | *(College of Mathematics and Computer Science,Quanzhou Normal University,Quanzhou 362000,China) |
| |
| Abstract: | A novel Fuzzy clustering Algorithm based on Mahalanobis distance is proposed,and the update formulas are devised.This algorithm has a more explicable objective function and is more effective in computing.Experiments on artificial data set and cucumber data set verifies the effectiveness of this new algorithm.7 hue statistics of cucumber leaves are measured,and the semi supervised clustering algorithm is perform on a 2 dimension data that is deduced from the cucumber data by linear discriminant analysis.The correct recognition rata is 100% on labeled data,and above 96% on unlabeled data. |
| |
| Keywords: | Mahalanobis distance semi supervised fuzzy clustering algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
