用于高斯混合模型参数估计的EM算法及其初始化研究

基于有限混合模型的聚类是一种重要的聚类分析方法,而EM算法是混合模型参数估计的重要方法.传统的EM算法对初始聚类中心比较敏感,因此如何选取初始值成为运用EM算法实现高斯混合模型聚类中的一个重要问题.本文提出一种基于网格的聚类算法来初始化EM算法,旨在改善EM算法的初始敏感性,使其达到更佳的聚类效果.此算法根据网格单元密...

EM algorithm and its initialization research for parameter estimation of Gaussian mixture models

Clustering based on finite mixture models is a kind of important clustering analysis methods,and the EM algorithm is an important method of parameter estimation of mixture models.The traditional EM algorithm is sensitive to initial clustering center,and therefore how to choose initial values has become an important problem in realizing clustering based on Gaussian mixture models using the EM algorithm.In this paper,a clustering algorithm based on grid to initialize the EM algorithm has been put forward that...