Link

The links of a Bayes net indicate which nodes (i.e. variables) are directly dependent on others.  More accurately, they indicate which nodes are independent of which others.  If there is no link from one node to another, then their corresponding variables are conditionally independent given values for their parents.  Using the d-separation rule, you can determine which nodes effect which others during probabilistic inference, based on the link structure link of the net.

A link between two nodes is indicated as a line between them with an arrow at one end, there are several ways to add links between nodes.

A link into a decision node is known as an informational link, and indicates that the decision maker will know the values of the parent nodes when a decision must be made.

Some people refer to links as “arcs” or “edges”.