|
|||||
|
|
| 基于最大惩罚似然的高斯混合模型无监督分类研究 | |
| 引用本文： | 余鹏,童行伟,封举富.基于最大惩罚似然的高斯混合模型无监督分类研究[J].应用概率统计,2008,24(5):475-483. |
| 作者姓名： | 余鹏 童行伟 封举富 |
| 作者单位： | 1. 北京大学数中图学学院,北京,1000871;国家基础地理信息中心,北京,100044 2. 北京师范大学数中图学学院,北京,100875 3. 北京大学信息科学学院信息科学中心、视觉与听觉信息处理国家重点实验室,北京,100871 |
| 摘 要： | 本文提出了一个基于高斯混合模型的无监督分类算法. 考虑到利用EM算法求解高斯混合模型的参数参数估计问题容易陷入局部最优解, 我们引入逆Wishart分布来代替传统的Jeffery先验. 几个实验数据的结果表明, 采用该方法估计无监督分类的成分数, 无论是估计的正确率, 还是运算速度, 都有较大提高. |
| 关 键 词： | 高斯混合模型 无监督分类 最大惩罚似然 EM算法 逆Wishart分布. |
Unsupervised Classification Based on Penalized Maximum Likelihood of Gaussian Mixture Models |
|
| YU PENG,TONG XINWEI,FENG JUFU.Unsupervised Classification Based on Penalized Maximum Likelihood of Gaussian Mixture Models[J].Chinese Journal of Applied Probability and Statisties,2008,24(5):475-483. | |
| Authors: | YU PENG TONG XINWEI FENG JUFU |
| Institution: | School of Mathematical Sciences;Peking University;Beijing;100871;National Geomatics Center of China;100044;School of Mathematical Sciences;Beijing Normal University;100875;National Laboratory on Machine Perception;Center for Information Science;School of Electronics Engineering and Computer Science;100871 |
| Abstract: | In this paper we propose an unsupervised classification algorithm which is based on Gaussian mixture models. Thinking that EM algorithm will result in a local optimal resolution of Gaussian mixture models in parameter estimations, we substitute invert Wishart distribution for Jeffery prior. Experiments show that this algorithm improves correct rates and decreases time while estimating classifications. |
| Keywords: | Gaussian mixture models unsupervised Classification penalized maximum likelihood EM algorithm invert Wishart distribution |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《应用概率统计》浏览原始摘要信息 | |
| 点击此处可从《应用概率统计》下载免费的PDF全文 | |