7. Conclusions

These concerns aside, Bayesian networks have incredible power to offer assistance in a wide range of endeavours. They support the use of probabilistic inference to update and revise belief values. Bayesian networks readily permit qualitative inferences without the computational inefficiencies of traditional joint probability determinations. In doing so, they support complex inference modelling including rational decision making systems, value of information and sensitivity analysis. As such, they are useful for causality analysis and through statistical induction they support a form of automated learning. This learning can involve parametric discovery, network discovery, and causal relationship discovery.[21]

In this paper, we discussed the premises of Bayesian networks from Bayes' Theorem and how such Bayesian inference differs from classical treatments of statistical inference. The reasons, implications and emerging potential of Bayesian networks in the area of Artificial Intelligence were then explored with an applied focus profiling some current areas where Bayesian networks and models are being employed to address real-life problems. Finally, we examined some of the limitations of Bayesian networks.

At best, such a paper can only be a snapshot of the state of Bayesian research at a given time and place. The breadth and eclectic foci of the many individuals, groups and corporations researching this topic makes it one of the truly dynamic areas within the discipline of Artificial Intelligence.