2.2.4.2.2.16 MiningModel
This complex type MAY<93> represent a mining model.
-
<xsd:complexType name="MiningModel"> <xsd:all> <!--These elements are common to each MajorObject--> <xsd:element name="Name" type="xsd:string" /> <xsd:element name="ID" type="xsd:string" minOccurs="0" /> <xsd:element name="CreatedTimestamp" type="xsd:dateTime" minOccurs="0" /> <xsd:element name="LastSchemaUpdate" type="xsd:dateTime" minOccurs="0" /> <xsd:element name="Description" type="xsd:string" minOccurs="0" /> <xsd:element name="Annotations" minOccurs="0" > <xsd:complexType> <xsd:sequence> <xsd:element name="Annotation" type="Annotation" minOccurs="0" maxOccurs="unbounded" /> </xsd:sequence> </xsd:complexType> </xsd:element> <!--Extended elements for MiningModel object--> <xsd:element name="Algorithm" minOccurs="1"> <xsd:simpleType> <xsd:restriction base="xsd:string" > <xsd:enumeration value="Microsoft_Naive_Bayes" /> <xsd:enumeration value="Microsoft_Decision_Trees" /> <xsd:enumeration value="Microsoft_Clustering" /> <xsd:enumeration value="Microsoft_Neural_Network" /> <xsd:enumeration value="Microsoft_Logistic_Regression" /> <xsd:enumeration value="Microsoft_Linear_Regression" /> <xsd:enumeration value="Microsoft_Association_Rules" /> <xsd:enumeration value="Microsoft_Time_Series" /> <xsd:enumeration value="Microsoft_Sequence_Clustering" /> </xsd:restriction> </xsd:simpleType> </xsd:element> <xsd:element name="LastProcessed" type="xsd:dateTime" minOccurs="0"/> <xsd:element name="AlgorithmParameters" minOccurs="0"> <xsd:complexType> <xsd:sequence> <xsd:element name="AlgorithmParameter" type="AlgorithmParameter" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="AllowDrillThrough" type="xsd:boolean" minOccurs="0"/> <xsd:element name="Translations" minOccurs="0"> <xsd:complexType> <xsd:sequence> <xsd:element name="Translation" type="AttributeTranslation" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="Columns" > <xsd:complexType> <xsd:sequence> <xsd:element name="Column" type="MiningModelColumn" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="State" minOccurs="0"> <xsd:simpleType> <xsd:restriction base="xsd:string" > <xsd:enumeration value="Processed" /> <xsd:enumeration value="Unprocessed" /> </xsd:restriction> </xsd:simpleType> </xsd:element> <xsd:element name="FoldingParameters" type="FoldingParameters" minOccurs="0" /> <xsd:element name="Filter" type="xsd:string" minOccurs="0" /> <xsd:element name="MiningModelPermissions" minOccurs="0"> <xsd:complexType> <xsd:sequence> <xsd:element name="MiningModelPermission" type="MiningModelPermission" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="Language" type="xsd:string" minOccurs="0"/> <xsd:element name="Collation" type="xsd:string" minOccurs="0"/> </xsd:all> </xsd:complexType>
The following table describes the elements that are included in the XSD schema for MiningModel. Those elements common to all major objects are described in section 2.2.4.2.2.1.
Element |
Read-Only |
Default value |
Description |
---|---|---|---|
Algorithm |
|
[Required] |
The protocol does not require any particular algorithms to be supported, and each server developer can support whichever algorithms he or she chooses to support.<94> |
LastProcessed |
Yes |
|
The date and time when the mining model was last processed. |
AlgorithmParameters |
|
Empty |
A collection of objects of type AlgorithmParameter. The allowed parameters are different depending on the algorithm. |
AllowDrillThrough |
|
False |
When true, indicates that drillthrough is allowed; otherwise, false. |
Translations |
|
Empty |
A collection of Translation objects. |
Columns |
|
Empty |
A collection of objects of type MiningModelColumn. |
State |
Yes |
|
Represents the processing state of the partition. Values include:
|
FoldingParameters |
|
Empty |
An object of type FoldingParameters. Describes a fold (a partition of the training data) to be used for training this mining model. Used only as part of the multifold cross-validation procedure.<95> |
Filter |
|
Empty |
The DMX filter statement to be applied to training data for models that are trained only on a part of a structure's data. An empty string or missing element implies no filter.<96> |
MiningModelPermissions |
|
Empty |
A collection of MiningModelPermission objects. |
Language |
|
Empty |
The language to use by default. |
Collation |
|
Empty |
The collation sequence to use. |