Home > Data Mining Algorithms > Anomaly Detection > Anomaly Detection Viewers a... > Association > AR Model Viewers and Algori... > Decision Tree > Expectation Maximization > EM Model Viewer and Algorit... > Generalized Linear Models > GLM Model Viewers and Algor... > GLM Classification Algorith... > Feature Selection Option Di...
Requires connection to Oracle Database 12c.
If you select Feature Selection, Ridge Regression is automatically de-selected.
This dialog allows you to specify feature selection for a GLMC or a GLMR model:
Feature Selection Criteria: The default is System Determined.
You can explicitly select one of the following:
Akaike Information
Schwarz Bayesian Information
Risk Inflation
Alpha Investing
Max Number of Features: The default is System Determined.
To specify a number of features, click the radio button User specified and enter an integer number of features.
Feature Identification: The default is System Determined.
You can also choose
Enable Sampling
Disable Sampling
Feature Acceptance: The default is System Determined.
You can also choose
Strict
Relaxed
Prune Model: The default is Enable; you can also select Disable.
Categorical Predictor Treatment: The default is Add One at a Time; You can also select Add All at Once.
If you accept the default (Add One at a Time), the default is to not select Feature Generation. If you check (select) Feature Generation, the default is Quadratic Candidates; you can also select Cubic Candidates.