| 基于惩罚距离的混合模型分量数自动估计算法 |
| |
| 引用本文： | 张大明,符茂胜,郭慧,罗斌.基于惩罚距离的混合模型分量数自动估计算法[J].华南理工大学学报(自然科学版),2009,37(10). |
| |
| 作者姓名： | 张大明 符茂胜 郭慧 罗斌 |
| |
| 作者单位： | 1. 安徽大学,计算机科学与技术学院,安徽,合肥,230039
2. 新南威尔士大学,澳大利亚,悉尼,NSW2052 |
| |
| 基金项目： | 国家自然科学基金资助项目，高等学校博士学科点专项科研基金资助项目，安徽省高等学校自然科学研究重点项目 |
| |
| 摘 要： | EM算法是对有限混合模型进行参数估计的通用算法, 然而标准EM算法中所需的混合模型分量数实际上往往是未知的. 研究并改进了一种采用惩罚性最小匹配距离估计分量数的方法, 结合贪婪EM算法框架, 提出一种可以在进行参数估计的同时, 快速准确地自动估计高斯混合模型分量数的算法. 通过一元和二元的高斯混合模型的仿真实验, 验证了算法的有效性.
关键词：有限混合模型; 分量数; 惩罚性最小匹配距离; 贪婪EM; Parzen窗; 带宽
|
| 关 键 词： | 有限混合模型 分量数 惩罚性最小匹配距离 贪婪EM Parzen窗 带宽 |
| 收稿时间： | 2008-11-20 |
| 修稿时间： | 2009-3-25 |
An Automatic Algorithm for Estimating the Number of Components of Gaussian Mixture by Penalized Distance |
| |
| Zhang Da-ming,Fu Mao-sheng,Guo Hui,Luo Bin.An Automatic Algorithm for Estimating the Number of Components of Gaussian Mixture by Penalized Distance[J].Journal of South China University of Technology(Natural Science Edition),2009,37(10). |
| |
| Authors: | Zhang Da-ming Fu Mao-sheng Guo Hui Luo Bin |
| |
| Abstract: | The expectation-maximization (EM) algorithm is a popular approach for parameter estimation of finite mixture model (FMM). A drawback of this approach is that the number of components of the finite mixture model is not known in advance, nevertheless, it is a key issue for EM algorithms. In this paper, a penalized minimum matching distance-guided EM algorithm is discussed. Under the framework of Greedy EM, an automatic, fast and accurate algorithm for estimating the number of components of the Gaussian mixture model (GMM) is proposed. The performance of this algorithm is validated via simulative experiments of univariate and bivariate Gaussian mixture models. |
| |
| Keywords: | finite mixture model number of components penalized minimum matching distance greedy EM Parzen window bandwidth |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《华南理工大学学报(自然科学版)》浏览原始摘要信息 |
|
