Microsoft SQL Server Decision Tree Training System

References

http://technet.microsoft.com/en-us/library/cc645868.aspx
- The Microsoft Decision Trees algorithm is a hybrid algorithm that incorporates different methods for creating a tree, and supports multiple analytic tasks, including regression, classification, and association. The Microsoft Decision Trees algorithm supports modeling of both discrete and continuous attributes.
  This topic explains the implementation of the algorithm, describes how to customize the behavior of the algorithm for different tasks, and provides links to additional information about querying decision tree models.
  ... The Microsoft Decision Trees algorithm applies the Bayesian approach to learning causal interaction models by obtaining approximate posterior distributions for the models.

http://technet.microsoft.com/en-us/library/cc645758.aspx
- This topic describes mining model content that is specific to models that use the Microsoft Decision Trees algorithm. For a general explanation of mining model content for all model types, see Mining Model Content (Analysis Services - Data Mining). It is important to remember that The Microsoft Decision Trees algorithm is a hybrid algorithm that can create models with very different functions: a decision tree can represent associations, rules, or even linear regression. The structure of the tree is essentially the same, but how you interpret the information will depend on the purpose for which you created the model.