Data Mining with Microsoft® SQL Server® 2008

Appendix B. Supported Functions

This appendix lists (as a reference) all the functions supported by DMX. The first section covers the native DMX language functions. The next two sections list the Visual Basic for Applications (VBA) and Excel functions that can be called as stored procedures from the DMX language. The last section of this appendix describes supplemental stored procedures for creating and deleting data set objects.

DMX Language Functions

These functions are natively supported by the DMX language. Each individual function may or may not be supported by any particular algorithm. Table B-1 lists the functions that operate on scalar columns and Table B-2 lists functions that operate on table columns.

VBA Functions

Table B-3 lists the supported VBA functions.

Excel Functions

Table B-4 lists the supported Excel functions.

Table B.1. Prediction Functions on Scalar Columns

FUNCTION	RETURN VALUE	DESCRIPTION
Cluster()	Cluster identifier	Cluster identifier that the input case belongs to with the highest probability. It also can be used as a <cluster column reference> for a PredictHistogram function.
ClusterDistance ([Cluster Identifier])	Continuous value	Distance from the center of the cluster that is identified by Cluster Identifier or the highest probability cluster. Typically this value is 1-ClusterProbability().
ClusterProbability ([Cluster Identifier])	Probability (0-1) of cluster membership	Probability that the input case belongs to the cluster that is identified by Cluster Identifier or ...