SELECT FROM <model>.CONTENT (DMX)
Returns the mining model schema rowset for the specified data mining model.
Syntax
SELECT [FLATTENED] [TOP <n>] <expression list> FROM <model>.CONTENT
[WHERE <condition expression>]
[ORDER BY <expression> [DESC|ASC]]
Arguments
- n
Optional. An integer specifying how many rows to return.
- expression list
A comma-separated list of columns derived from the Content schema rowset.
- model
A model identifier.
- condition expression
Optional. A condition to restrict the values that are returned from the column list.
- expression
Optional. An expression that returns a scalar value.
Remarks
The SELECT FROM <model>.CONTENT statement returns content that is specific to each algorithm. For example, you may want to use the descriptions of all the rules of an association rules model in a custom application. You can use a SELECT FROM <model>.CONTENT statement to return values in the NODE_RULE column of the model.
The following table provides a general description of the columns that are included in the mining model content.
Note
Algorithms may interpret the columns differently in order to properly represent the content.
| CONTENT rowset column | Description |
|---|---|
MODEL_CATALOG |
A catalog name. NULL if the provider does not support catalogs. |
MODEL_SCHEMA |
An unqualified schema name. NULL if the provider does not support schemas. |
MODEL_NAME |
A model name. This column cannot contain a NULL. |
ATTRIBUTE_NAME |
The names of the attributes that correspond to this node. |
NODE_NAME |
The name of the node. |
NODE_UNIQUE_NAME |
A unique name of the node. |
NODE_TYPE |
The type of the node, which is algorithm-specific. |
NODE_GUID |
The node GUID. NULL if no GUID. |
NODE_CAPTION |
A label or a caption that is associated with the node. Used primarily for display purposes. If a caption does not exist, NODE_NAME is returned. |
CHILDREN_CARDINALITY |
The number of children that the node has. |
PARENT_UNIQUE_NAME |
A unique name of the node's parent. NULL is returned for any nodes at the root level. |
NODE_DESCRIPTION |
A readable description of the node. |
NODE_RULE |
An XML description of the rule embedded in the node. The format of the XML string is based on the PMML standard. |
MARGINAL_RULE |
An XML description of the rule moving to the node from the parent node. |
NODE_PROBABILITY |
The probability for reaching the node. |
MARGINAL_PROBABILITY |
The probability of reaching the node from the parent node. |
NODE_DISTRIBUTION |
A table containing the probability histogram of the node. |
NODE_SUPPORT |
The number of cases in support of this node. |
For More Information:Data Mining Schema Rowsets
Examples
The following code returns all the columns that define the Target Mail mining model for nodes that are descended from the node identified by '00000000100'.
SELECT * FROM [TM Decision Tree].CONTENT
WHERE ISDESCENDANT('00000000100')
See Also
Reference
SELECT (DMX)
Data Mining Extensions (DMX) Data Manipulation Statements
Data Mining Extensions (DMX) Statement Reference