Netica can do extensive utility-free single-finding sensitivity analysis. Select a node (called the "target node") and choose Network → Sensitivity to Findings from the menu. A report will be displayed in the Messages window displaying how much the beliefs, mean value, etc. of the target node could be influenced by a single finding at each of the other nodes in the net (each is called a "findings nodes").
The first part of the report has a section for each findings node, showing how much it can effect the target node using several different sensitivity measures. The second part is a summary table which compares the sensitivities for each of the findings nodes.
If you want to limit the report to a few findings nodes, first select the target node, and then use ctrl-select to add the desired findings nodes to the selection. Then choose Network → Sensitivity to Findings.
Currently this sensitivity analysis will only work for Bayes nets and not decision nets.
Example: Suppose you are using the net for diagnosis, and you want to determine which test is going to provide the best information about the presence of a fault or disease. Select the node for the fault or disease and choose Sensitivity to Findings. Use the summary list of sensitivities at the end of the report generated to identify possible findings nodes which will provide the most information about the fault/disease node. If you want more detailed information of how these findings nodes can effect the fault/disease node, look up each of them in the first part of the report.
Single Number: If you want a single number which best describes the degree of sensitivity of one node to another, it is recommended that you use the first column of the summary report at the end. For continuous nodes or nodes with state values defined, this will be the variance reduction, otherwise it will be the mutual information (i.e. entropy reduction).
Findings: When the sensitivities are calculated, the findings currently entered into the net will be taken into account, which can effect the sensitivities significantly.
For full documentation on this function, and each of the
sensitivity measures calculated,see
Sensitivity to Findings.