The HMM parameters were estimated by the EM algorithm.
隐马尔可夫模型参数通过期望最大化算法(EM)来估计。
EM algorithm has become one of the methods of choice for ML estimation.
EM算法是一种很有效的最大似然估计方法。
The EM algorithm is used to cluster traffics with interactive features.
用EM算法研究了具有交互特征的网络流量的分类；
The introduction of the improved EM algorithm also reduces the risk of data underflow .
使用一种改进的EM算法降低了数据下溢的风险。
Recently, the progress has been made on the research of the EM algorithm for Gaussian mixtures.
近年来，对于高斯混合体em算法的收敛性研究有了新的进展。
Firstly, we introduce the theory of finite mixture model and EM algorithm for maximum likelihood estimation.
首先，介绍了有限混合模型理论及应用EM算法求解极大似然估计。
We further obtain and prove the condition of the correct convergence of the EM algorithm for Gaussian mixtures.
理论分析和数值实验结果表明，高斯混合密度的EM算法的正确收敛性与混合密度的重叠度密切相关。
After analysis of em algorithm, we presented a new cooperative training algorithm based on incremental learning.
本文在分析了EM算法的基础上，提出了一种新的协同训练算法。
We describe the maximum likelihood parameter estimation problem and how the em algorithm can be used for its solution.
描述最大似然参数估计问题，介绍如何用EM算法求解最大似然参数估计。
METHODS Based on correlation information among data, the authors analyzed data by using EM algorithm and growth curve model.
方法通过数据间的相关信息，应用EM算法和生长曲线模型进行数据分析。
After explaining the EM algorithm, this paper gives the derivation of the multi-user detection algorithm based on the EM method.
在讨论了EM算法的基本原理之后，本文详细推导了基于EM方法的多用户检测算法。
It USES Gassian mixture model to represent particles and adopts EM algorithm to refit particles after correction step at each time.
该算法使用混合高斯模型表示粒子，在每个时刻的修正步骤之后，采用EM算法对粒子进行重新拟合。
The estimation of the parameters can be easily done through EM algorithm and the order model is also easily selected by BIC criterion.
给出了该模型参数估计的EM算法，并利用BIC准则对模型进行定阶。
We calculate the ML estimation via the EM algorithm, and derive its iteration equations, which gives a closed-form solution for parameters.
我们基于EM算法来计算参数的ML估计，推导了对应的参数迭代方程，给出了参数的一个闭式解。
Especially, we give some results of the convergence of the EM algorithm for the curved exponential family under the conditions checked easily.
特别对应用广泛的曲指数族，本文在较易实际验证的条件下给出了相应的EM算法的收敛结果。
In this paper we discuss the convergence of the EM algorithm for iterative computation of maximum likelihood estimates when the observations can be viewed as incomplete data.
本文讨论EM算法的收敛性，其中EM算法是不完全数据处理中的一类重要的参数估计的迭代算法。
First we introduce the abstract form of the EM algorithm. Then we develop the EM parameter estimation procedure for one application: finding the parameters of a mixture of Gaussian densities.
首先给出em算法的抽象形式，然后研究EM参数估计方法的一个应用：求高斯混合密度的参数。
In this paper, the defects latent semantic analysis, probabilistic latent semantic analysis using methods to construct the text-the words of co-occurrence matrix, using the em algorithm to solve.
本文针对潜在语义分析存在的缺陷，采用概率潜在语义分析的方法构造文本——词语的同现矩阵，使用EM算法进行迭代求解。
This article proposes a data sorting method via the EM algorithm, for the purpose of mining high-quality decisions by performing data reasoning in a database with incomplete, noisy and uncertain data.
针对存在不完整、含噪声和不确定数据的数据库，通过挖掘高质量的决策，对数据库的数据进行推理，提出了一种基于EM算法的数据清理方法。
The algorithm USES the Expectation Maximization (EM) clustering method to identify clusters and their sequences.
该算法采用期望最大化(EM)聚类分析方法来识别分类及其顺序。
We resort to expectation maximization (EM) algorithm for both the estimation of model parameters and the coping with missing values.
这里，期望最大化算法既用来处理丢失值又用来估计模型参数。
This algorithm can not only keep the merits of the original EM, but also facilitate the results converge o the global minimum.
该算法既保持了原EM算法的优点，又有利于训练结果收敛到全局极小。
To improve the accuracy of tracking the complex maneuver target in cluttered environment, a new state estimation algorithm based on the expectation maximization (EM) algorithm is presented.
为了提高在杂波环境下跟踪强机动目标的精度，提出了一种新的基于期望极大化(EM)算法的机动目标状态估计方法。
The multi-user detection algorithm based on the EM method is used to look for the maximum-likelihood estimation of users' data iteratively in a DS-CDMA system.
基于EM方法的多用户检测算法采用EM迭代方法来求解DS-CDMA系统中各用户发送数据的最大似然估计解。
To overcome the overflow difficulty existing in HMT model, a scaling algorithm is developed to improve expectation maximization (EM) algorithm.
为了克服HMT模型存在的计算溢出困难，采用尺度变换对EM算法进行了改进。
The deconvolved results with the compensated data and the original image data by expectation maximization (EM) algorithm for reducing the effect of out of focus light were compared respectively.
在此基础上给出期望最大化算法图像恢复结果，并对恢复结果做出分析。
The deconvolved results with the compensated data and the original image data by expectation maximization (EM) algorithm for reducing the effect of out of focus light were compared respectively.
在此基础上给出期望最大化算法图像恢复结果，并对恢复结果做出分析。
应用推荐