Mining Model Columns

Applies to: SQL Server 2019 and earlier Analysis Services Azure Analysis Services Fabric/Power BI Premium

Important

Data mining was deprecated in SQL Server 2017 Analysis Services and now discontinued in SQL Server 2022 Analysis Services. Documentation is not updated for deprecated and discontinued features. To learn more, see Analysis Services backward compatibility.

A data mining model applies a mining model algorithm to the data that is represented by a mining structure. Like the mining structure, the mining model contains columns. A mining model is contained within the mining structure, and inherits all the values of the properties that are defined by the mining structure. The model can use all the columns that the mining structure contains or a subset of the columns.

You can define two additional pieces of information on a mining model column: usage, and modeling flags.

  • Usage is a property that defines how the model uses the column. Columns can be used as input columns, key columns, or predictable columns.

  • Modeling flags provide the algorithm with additional information about the data that is defined in the case table, so that the algorithm can build a more accurate model. You can define modeling flags programmatically by using the Data Mining Extensions (DMX) language, or in Data Mining Designer in SQL Server Data Tools.

The following list describes the modeling flags that you can define on a mining model column.

MODEL_EXISTENCE_ONLY
Indicates that the presence of the attribute is more important than the values that are in the attribute column. For example, consider a case table that contains a list of order items that are associated with a particular customer. The table data includes the product type, ID, and cost of each item. For modeling purposes, the fact that the customer purchased a particular order item may be more important than the cost of the order item itself. In this case, the cost column should be marked as MODEL_EXISTENCE_ONLY.

REGRESSOR
Indicates that the algorithm can use the specified column in the regression formula of regression algorithms. This flag is supported by the Microsoft Decision Trees and Microsoft Time Series algorithms.

For more information about setting the usage property and defining modeling flags programmatically with DMX, see CREATE MINING MODEL (DMX). For more information about setting the usage property and defining modeling flags in SQL Server Data Tools, see Moving Data Mining Objects.

See Also

Data Mining Algorithms (Analysis Services - Data Mining)
Mining Structures (Analysis Services - Data Mining)
Change the Properties of a Mining Model
Exclude a Column from a Mining Model
Mining Structure Columns