Data Mining Stored Procedures (Analysis Services - Data Mining)
Beginning in SQL Server 2005, Analysis Services supports stored procedures that can be written in any managed language. The managed languages that are supported include Visual Basic .NET, C#, and managed C++. In SQL Server Management Studio, you can call the stored procedures directly by using the CALL statement, or as part of a Data Mining Extensions (DMX) query.
For more information about calling Analysis Services stored procedures, see Calling Stored Procedures.
For general information about the programmability of SQL Server 2012 Analysis Services (SSAS), see Data Mining Programming.
For additional information about how to program data mining objects, see the article, "SQL Server Data Mining Programmability", in the MSDN library.
When you query mining models, especially when you test new data mining solutions, you might find it convenient to call the system stored procedures that are used internally by the data mining engine. You can view the names of these system stored procedures by using SQL Server Profiler to create a trace on the Analysis Services server, and then creating, browsing, and querying the data mining models. However, Microsoft does not guarantee the compatibility of system stored procedures between versions, and you should never use calls to the system stored procedures in a production system. Instead, for compatibility, you should create your own queries by using DMX or XML/A.