tiberius data mining Predictive Modelling Software 'attractively simple'   
  Tiberius Data Mining





1 . Remove Redundant VariablesRemove Redundant Variables

Remove Redundant Variables

After the logistic model is built, variables are tested for importance. Selecting this option will remove unimportant variables and repeat the model building process. This option is recommended as it leads to more robust models.


2 . Validation SetValidation Set

Validation Set

A random percentage of the data can be withheld from the training phase and used as an independent validation set to determine if the final model has good generalisation properties.


3 . Sub PopulationSub Population

Sub Population

The model can be built on specified sub-populations of the data. Use this frame to specify any sub-population to be used.


4 . Build the ModelBuild the Model

Build the Model

Once everything is set, press this button to build the model.