A modeling technique that provides a mathematically sound formalism for representing and reasoning about uncertainty, imprecision, or unpredictability in our knowledge. For example, seeing that the front lawn is wet, one might wish to determine whether it rained during the previous night. Inference algorithms can use the structure of the Bayesian network to calculate conditional probabilities based on whatever data has been observed (e.g., the street does not appear wet, so it is 90% likely that the wetness is due to the sprinklers). Bayesian networks offer or enable a set of benefits not provided by any other system for dealing with uncertainty - an easy to understand graphical representation, a strong mathematical foundation, and effective automated tuning mechanisms. These techniques have proved useful in a wide variety of tasks including medical diagnosis, natural language understanding, plan recognition, and intrusion detection. Also called belief networks, Bayes networks, or causal probabalistic networks.

This definition is in context to Artificial Intelligence. See more contextual defintions for Bayesian Networks.