Setting up and running RT classification

To run an RT classification, you must have a training data set for Object Analyst to learn from and create a classification model that is used to predict the class membership of the population.

To set up and run an RT classification

In the Object Analyst window, in the Maximum tree depth box, type or select a value according to the maximum permissible depth of the tree.
This parameter specifies the maximum number of levels leaf nodes below the root node. The training algorithms attempt to split a node while its depth is less than the value you specify. The depth of the optimal tree may be less than you specify if the other termination criteria are met before the maximal depth is reached.

A small value will likely result in a tree with a large variance while, conversely, a large value will tend to overfit the training model.
In the Minimum samples count box, type or select a value according to the minimum number of samples at a leaf node to allow it to be further split into child nodes.
A reasonable value is a small percentage of the total number of training samples, such as, one percent. For example, if the attribute table contains 2000 training samples (_T), a reasonable value would be 20.
In the Active variables box, type or select a value according to the number of randomly selected subset of attributes at each tree node to decide if that node should be split.
Set to 0 (zero) by default, the size is set to the square root of the total number of attributes.
In the Termination criteria list, click one of the following to determine how you want to stop the training algorithm:
- Both uses both termination criteria.
- Trees accuracy ends learning when the trees accuracy is reached.
  Type or select the accuracy value you want to use in the Trees accuracy box, below.
- Maximum number of trees ends learning when the maximum number of trees in the forest is reached.
  Type or select the maximum number of trees you want to use in the Maximum number of trees box, below.
Return to Setting up and running Supervised Classification.