After a Bayes net has been constructed and compiled, you can apply it to a particular case by entering findings (also known as “evidence”), which are the values for nodes that you know. Then you use belief updating (which is a form of probabilistic inference) to determine new probabilities for the states of all the other nodes.
You can run through the process using this example.
As well as finding beliefs, Netica can find the most likely configuration of the remaining variables, according to the findings entered. This is called the “Most Probable Explanation”.
You can also use Netica to do sensitivity
analysis to find how tightly coupled nodes are. That can be
used to determine the degree to which a finding
at one node is expected to change the beliefs at other nodes, or which
nodes would be the best to obtain findings for, in order to obtain maximum
information on another node.