Content Types (DMX)

Data mining algorithms require additional information beyond the data type to function correctly, such as the content type. The content type helps the algorithm determine how to work with the data in the column.

Each algorithm supports specific content types. For example, the Microsoft Naive Bayes algorithm cannot use continuous columns. To use a continuous column in a Microsoft Naive Bayes model, you must discretize the data in the column. Some algorithms require certain content types in order to function correctly. For example, the Microsoft Time Series algorithm requires a key time column to identify the time over which the data was collected.

For a complete description of the content types that Analysis Services supports, see Content Types (Data Mining).