Automatically Building Diagnostic Bayesian Networks from On-line Data Sources and the SMILE Web-based Interface
Anucha Tungkasthan, Nipat Jongsawat, Pittaya Poompuang, Sarayut Intarasema, Wichian Premchaiswadi · InTech eBooks · 2010
data sources in more detail.Section 5 presents a conclusion and discusses some perspectives and ideas for future work.An acknowledgement is described in section 6. FundamentalsThis section is intended to describe the fundamentals of Bayesian networks and the core reasoning engines of SMILE web-based interface development.They are described in the following sections. Bayesian network and Bayesian updatingBayesian networks (also called belief networks, Bayesian belief networks, causal probabilistic networks, or causal networks) (Pearl, 1988) are acyclic directed graphs in which nodes represent random variables and arcs represent direct probabilistic dependencies among them.The structure of a Bayesian network is a graphical, qualitative illustration of the interactions among the set of variables that it models.The structure of the directed graph can mimic the causal structure of the modeled domain, although this is not necessary.When the structure is causal, it provides a useful, modular insight into the interactions among the variables and allows for a prediction of the effects of external manipulation.Nodes of a Bayesian network are usually drawn as circles or ovals.A Bayesian network also represents the quantitative relationships among the modeled variables.Numerically, it represents the joint probability distribution among them.This distribution is described efficiently by exploring the probabilistic independence among the modeled variables.Each node is described by a probability distribution conditional on its direct predecessors.Nodes with no predecessors are described by prior probability distributions.Both the structure and the numerical parameters of a Bayesian network can be elicited from an expert.They can also be derived from data, as the structure of a Bayesian network is simply a representation of interdependencies in the data and the numbers are a representation of the joint probability distributions that can be inferred from the data.Finally, both the structure and the numerical probabilities can be a mixture of expert knowledge, measurements and objective frequency data.Bayesian updating, also referred to as belief updating, or somewhat less precisely as probabilistic inference is based on the numerical parameters captured in the model (Cooper, 1990).The structure of the model which is an explicit statement of the interdependencies in the domain helps in making the algorithms for Bayesian updating more efficient (Dagum & Luby, 1997).All algorithms for Bayesian updating are based on a theorem proposed by Rev. Thomas Bayes (1702-1761) and are known as Bayes Theorem.Belief updating in Bayesian networks is computationally complex.In the worst case, belief updating algorithms are NPhard (Cooper 1990).There exist several efficient algorithms, however, that make belief updating in graphs consisting of tens or hundreds of variables tractable.Pearl developed a message-passing scheme that updates the probability distributions for each node in a Bayesian network in response to observations of one or more variables (Pearl, 1986).Lauritzen and Spiegelhalter, Jensen et al, and Dawid proposed an efficient algorithm that first transforms a Bayesian network into a tree where each node in the tree corresponds to a subset of variables in the original graph (Lauritzen & Spiegelhalter, 1988: Jensen et al., 1990: Dawid, 1992).The algorithm then exploits several mathematical properties of this tree to perform probabilistic inference.Several approximate algorithms based on stochastic www.intechopen.com How to referenceIn order to correctly reference this scholarly work, feel free to copy and paste the following: Anucha Tungkasthan, Nipat Jongsawat, Pittaya Poompuang, Sarayut Intarasema and Wichian Premchaiswadi (2010).Automatically Building Diagnostic Bayesian Networks from On-line Data Sources and the SMILE Webbased Interface, Decision Support Systems, Chiang S. Jao (Ed.),