The threshold EM algorithm for parameter learning in bayesian network with incomplete data

被引：2

作者：

Ben Lamine, Fradj ^{[1
]}

Kalti, Karim ^{[1
]}

Mahjoub, Mohamed Ali ^{[2
]}

机构：

[1] Fac Sci Monastir, Dept Comp Sci, Monastir, Tunisia

[2] Preparatory Inst Engineer Monastir, Monastir, Tunisia

来源：

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS | 2011年 / 2卷 / 07期

关键词：

bayesian network; parameter learning; missing data; EM algorithm; Gibbs sampling; RBE algorithm; brain tumor;

D O I：

暂无

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

Bayesian networks (BN) are used in a big range of applications but they have one issue concerning parameter learning. In real application, training data are always incomplete or some nodes are hidden. To deal with this problem many learning parameter algorithms are suggested foreground EM, Gibbs sampling and RBE algorithms. In order to limit the search space and escape from local maxima produced by executing EM algorithm, this paper presents a learning parameter algorithm that is a fusion of EM and RBE algorithms. This algorithm incorporates the range of a parameter into the EM algorithm. This range is calculated by the first step of RBE algorithm allowing a regularization of each parameter in bayesian network after the maximization step of the EM algorithm. The threshold EM algorithm is applied in brain tumor diagnosis and show some advantages and disadvantages over the EM algorithm.

引用

页码：86 / 91

页数：6

共 11 条

[1]

Basilio Sierra I. I., 2000, MED BAYES NETWORKS, DOI [10.1007/3-540-39949-6_2, DOI 10.1007/3-540-39949-6_2]

[2] Bayesian networks and information retrieval:: an introduction to the special issue [J].

de Campos, LM ;

Fernández-Luna, JM ;

Huete, JF .

INFORMATION PROCESSING & MANAGEMENT, 2004, 40 (05) :727-733

[3] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].

DEMPSTER, AP ;

LAIRD, NM ;

RUBIN, DB .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38

[4] STOCHASTIC RELAXATION, GIBBS DISTRIBUTIONS, AND THE BAYESIAN RESTORATION OF IMAGES [J].

GEMAN, S ;

GEMAN, D .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1984, 6 (06) :721-741

[5]

Jensen F. V., 1996, INTRO BAYESIAN NETWO

[6]

Lauritzen S., 1991, TR9105 DEP STAT AALB

[7]

Marco Ramoni P. S., 1997, USE EXOGENOUS KNOWLE

[8]

Marco Ramoni P. S., 1997, BAYESIAN INFERENCE M

[9] Growing Bayesian network models of gene networks from seed genes [J].

Peña, JM ;

Björkegren, J ;

Tegnér, J .

BIOINFORMATICS, 2005, 21 :224-229

[10] Robust learning with missing data [J].

Ramoni, M ;

Sebastiani, P .

MACHINE LEARNING, 2001, 45 (02) :147-170

← 1 2 →