Testing the randomness of the split with a SSAS decision trees model
You need to test the randomness of the split from the previous recipe. You will use the SSAS decision trees algorithm to check whether you can predict a set membership with input variables. You should get a very shallow tree, meaning there are no patterns that could explain the set membership.
Getting ready
For this recipe, you need to have SSAS installed in multidimensional and data mining mode. You also need to finish the previous recipe.
How to do it...
- In SSDT, add a new analysis services multidimensional and data mining project to the
Chapter08solution and name the projectCh08SSAS. - In solution explorer, right-click the
Data Sourcesfolder and selectNew Data Sourceto create a new data source. - In the
Data Source Wizardwelcome screen, clickNext. - In the
Selecthow to define the connection screen, create a data source using a connectionAdventureWorksDW2014database. ClickNext. - In the
Impersonation Informationscreen, you...
