GeNIe Tutorials: Tutorial 16 - Parameters
From DSL
Objective of the tutorial: To learn the parameters Bayesian network.
Estimated Time: 10 Minutes
At the end of the tutorial you will be able to:
- Edit data files to learn Bayesian networks.
To learn parameters of an existing network (structure already defined) you have to have a datafile and the network itself opened. Let's open retention_discretized.txt data file and param_learning.xdsl network. Then select Data -> Learn Parameters... from the main menu.
Next you need to create a mapping between the variables defined in the network (left column) and variables defined in the data set (right column). To do that simply drag-and-drop variables. All the established mappings will be shown in the middle column. You can further define the mapping of the states of those variables (also using drag-and-drop). For the purpose of this tutorial let's connect variables in the alphabetical order and do not change any states mappings. At this point the Learn Parameters window should look like that:
After you click Ok you can define the parameter confidence level and click Ok again to start the learning process.
Next Tutorial:
