Naive Bayes
Naive Bayes classifiers are among the most commonly used methods for supervised classification and they can be represented as Bayesian Networks. UnbBayes includes a wizard interface to guide the user through the definition of Bayesian network representing a Naive Bayes Classifier. This wizard interface can be evoked by using ...
When the command is issued, the wizard interface starts with the following screen.
Select data base
This step consists of choosing a database for the learning process using the Browse button.
Numeric attributes options
This panel defines the way in which continuous variables are handled in the current session.
The panel sets two preferences options:
Number of Discretized States |
The default number of intervals in which the function discretize from data will divide a continuous variable. |
Default Discretization Method |
UnbBayes provides two discretization methods and this item allows the user to choose the default method. The first method (Frequency) divides the range of values of a variable into intervals containing the same number of cases. The second method (Range) divides the range of values of a variable in equally spaced bins. |
Select class
This step consists of choosing the class variable.