CREATE MINING STRUCTURE (DMX)
Creates a new mining structure in a database. You can then use the ALTER MINING STRUCTURE (DMX) statement to add models to the mining structure.
You define a mining structure column by including the following information for each column in the column definition list:
Data type (mandatory)
List of modeling flags
Content type (mandatory)
Relationship to an attribute column (mandatory only if it applies), indicated by the RELATED TO clause
Use the following syntax for the column definition list, to define a single column:
<column name> <data type> [<Distribution>] [<Modeling Flags>] <Content Type> [<column relationship>]
Use the following syntax for the column definition list, to define a nested table column:
<column name> TABLE ( <non-table column definition list> )
Except for modeling flags, you can use no more than one clause from a particular group to define a column. You can define multiple modeling flags for a column.
For a list of the data types, content types, column distributions, and modeling flags that you can use to define a structure column, see the following topics:
Data Types (Data Mining)
Content Types (Data Mining)
Modeling Flags (Data Mining)
You can add a clause to the statement to describe the relationship between two columns. Analysis Services supports using the following <Column relationship> clause.
- RELATED TO
This form indicates a value hierarchy. The target of a RELATED TO column can be a key column in a nested table, a discretely-valued column in the case row, or another column with a RELATED TO clause, which indicates a deeper hierarchy.
The following example creates a mining structure called New Mailing, without creating any associated mining models. To learn how to add a mining model to the structure, see ALTER MINING STRUCTURE (DMX).
CREATE MINING STRUCTURE [New Mailing] ( CustomerKey LONG KEY, Gender TEXT DISCRETE, [Number Cars Owned] LONG DISCRETE, [Bike Buyer] LONG DISCRETE )