Bayesian Network

A probabilistic graphical model that represents a set of variables and their conditional dependencies via a directed acyclic graph.

How it works

A Bayesian network represents variables as nodes in a directed acyclic graph, with edges indicating causal or probabilistic dependencies. Each node stores a conditional probability table that quantifies the likelihood of its states given its parent nodes' states. Inference is performed by querying the network: given observed evidence (some variables' values), the algorithm propagates that information through the graph to compute the posterior probabilities of unobserved variables using Bayes' theorem.

Why it matters

Bayesian networks excel in domains where uncertainty must be modelled explicitly and causality matters. They are widely used in medical diagnosis, fault detection, and risk assessment. Unlike black-box neural networks, they provide interpretable probabilistic reasoning — you can trace exactly why a conclusion was reached. They were a dominant paradigm in AI before deep learning and continue to be valuable wherever data is scarce and expert knowledge can encode structure.