The threshold EM algorithm for parameter learning in bayesian network with incomplete data
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2011, Vol 2, Issue 7
Abstract
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.
Authors and Affiliations
Fradj Ben Lamine , Karim Kalti , Mohamed Ali Mahjoub
Keywords
Related Articles
- EP ID EP150357
- DOI -
- Views 86
- Downloads 0
Improving K-Means Algorithm by Grid-Density Clustering for Distributed WSN Data Stream
At recent years, Wireless Sensor Networks (WSNs) had a widespread range of applications in many fields related to military surveillance, monitoring health, observing habitat and so on. WSNs contain individual nodes that...
Assessing Assistive Learning Technologies with Experimental Design
Assistive learning technologies are generally computer-based instruments which are focused at supporting individuals with disabilities in enhancing their learning session with minimal intervention of parents, guardians,...
A Novel Intelligent Cluster-Head (ICH) to Mitigate the Handover Problem of Clustering in VANETs
The huge development in the number of Vehicle factories have resulted in many people having lost their life due to accident, which has made vehicular Ad-hoc networks (VANETs) hot topic to enable improved communication be...
Formal Specification of a Truck Geo-Location Big-Data Application
In the last few year’s social networks, e-commerce, mobile commerce, and sensor networks have resulted into an exponential increase in data size. This data comes in all formats i.e. structured, un-structured and semi-str...
Convex Hybrid Restoration and Segmentation Model for Color Images
Image restoration and segmentation are important areas in digital image processing and computer vision. In this paper, a new convex hybrid model is proposed for joint restoration and segmentation during the post-processi...