Share via


SystemGetClusterAccuracyResults (Analysis Services - Data Mining)

Returns cross-validation accuracy metrics for a mining structure and related clustering models.

This stored procedure returns metrics for the entire data set as a single partition. To partition the dataset into cross-sections and return metrics for each partition, use SystemGetClusterCrossValidationResults (Analysis Services - Data Mining).

Note

This stored procedure works only for clustering models. For non-clustering models, use SystemGetAccuracyResults (Analysis Services - Data Mining).

Syntax

SystemGetClusterAccuracyResults(
<mining structure> 
[,<mining model list>]
,<data set>
,<test list>])

Arguments

  • mining structure
    Name of a mining structure in the current database.

    (Required)

  • mining model list
    Comma-separated list of models to validate.

    The default is null, meaning that all applicable models are used. When the default is used, non-clustering models are automatically excluded from the list of candidates for processing.

    (Optional)

  • data set
    An integer value that indicates which partition in the mining structure is to be used for testing. The value is derived from a bitmask that represents the sum of the following values, where any single value is optional:

    Training cases

    0x0001

    Test cases

    0x0002

    Model filter

    0x0004

    For a complete list of possible values, see the Remarks section of this topic.

    (Required)

  • test list
    A string that specifies testing options. This parameter is reserved for future use.

    (optional)

Return Type

A table that contains scores for each individual partition and aggregates for all models.

The following table lists the columns returned by SystemGetClusterAccuracyResults. To learn more about how to interpret the information returned by the stored procedure, see Measures in the Cross-Validation Report.

Column Name

Description

ModelName

The name of the model that was tested. All indicates that the result is an aggregate for all models.

AttributeName

Not applicable to clustering models.

AttributeState

Not applicable to clustering models.

PartitionIndex

A number that indicates the partition.

For this stored procedure, the number is always 0.

PartitionCases

An integer that indicates how many cases have been tested.

Test

The type of test that was performed.

Measure

The name of the measure returned by the test. Measures for each model depend on the model type, and the type of the predictable value.

For a list of measures returned for each predictable type, see Measures in the Cross-Validation Report.

For a definition of each measure, see Cross-Validation (Analysis Services - Data Mining).

Value

A probability score that indicates the cluster case likelihood.

Remarks

The following table provides examples of the values that you can use to specify the data in the mining structure that is used for cross-validation. If you want to use test cases for cross-validation, the mining structure must already contain a testing data set. For information about how to define a testing data set when you create a mining structure, see Training and Testing Data Sets.

Integer Value

Description

1

Only training cases are used.

2

Only test cases are used.

3

Both the training cases and testing cases are used.

4

Invalid combination.

5

Only training cases are used, and the model filter is applied.

6

Only test cases are used, and the model filter is applied.

7

Both the training and testing cases are used, and the model filter is applied.

For more information about the scenarios in which you would use cross-validation, see Testing and Validation (Data Mining).

Examples

This example returns accuracy measures for two clustering models, named Cluster 1 and Cluster 2, that are associated with the vTargetMail mining structure. The code on line four indicates that the results should be based on the testing cases alone, without using any filters that might be associated with each model.

CALL SystemGetClusterAccuracyResults (
[vTargetMail],
[Cluster 1], [Cluster 2],
2
)

Sample Results:

ModelName

AttributeName

AttributeState

PartitionIndex

PartitionSize

Test

Measure

Value

Cluster 1

0

5545

Clustering

Case Likelihood

0.796514342249313

Cluster 2

0

5545

Clustering

Case Likelihood

0.732122471228572

Requirements

Cross-validation is available only in SQL Server Enterprise beginning in SQL Server 2008.

See Also

Reference

SystemGetCrossValidationResults (Analysis Services - Data Mining)

SystemGetAccuracyResults (Analysis Services - Data Mining)

SystemGetClusterCrossValidationResults (Analysis Services - Data Mining)

SystemGetClusterAccuracyResults (Analysis Services - Data Mining)