You are a SQL Server 2008 Analysis Services (SSAS) data mining architect. The customer table contains the following column names:
-Customer_key
-Name
-Age
-Education Level
-IsBuyer
You plan to build a data mining model by using Microsoft Decision Trees algorithm for the customer table.
You need to identify the data columnmodel parameter pairs to predict possible buyers.
Which model should you select?
A.
Data Column Model Parameter Type
Customer_key Input
Name Ignore
Education Level Input, Predict
Age Input, Predict
IsBuyer Key
B.
Data Column Model Parameter Type
Customer_key Key
Name Ignore
Education Level Input
Age Input
IsBuyer Predict
C.
Data Column Model Parameter Type
Customer_key Input
Name Ignore
Education Level Input
Age Input
IsBuyer Key
D.
Data Column Model Parameter Type
Customer_key Predict
Name Key
Education Level Input
Age Input
IsBuyer Input
Explanation:
Tip: "Microsoft Decision Trees … data columnmodel" = "IsBuyer Predict"http://technet.microsoft.com/en-us/library/ms175423.aspx
Changing Mining Column Usage
You can change which columns are included in a mining model and how each column is used, such as input, key, or predictable, by using the cells for that model column in the grid on the Mining Models tab. Each cell corresponds to a column in the mining structure. For key columns, you can set the cell to Key or Ignore. For input and output columns, you can set the cell to the following values:
– Ignore
– Input
– Predict
– PredictOnly
If you set a cell to Ignore, the column is removed from the mining model, but that column can still be used by other mining models in the structure.