Data Structure for Analytics

Oracle Business Intelligence repository contains the metadata that defines which columns (or piece of data) are available for you to include in analyses, and where data for each column originates. The repository is organized into subject areas, which contain folders with the columns.

Note: You can also use the Oracle BI repository as a data source for reports.

Columns

This table describes the three types of columns.

Column Type	Description	Example	Icon for Column Type
Fact	Provides a measure of something, meaning that the values are numbers.	Total	Yellow ruler
Attribute	Represents a piece of information about a business object, with values that are dates, IDs, or text.	Start Date	Gray paper
Hierarchy	Holds data values that are organized in a hierarchical manner.	Time, with sublevels: Year Quarter Month	Column: Hierarchy of blue squares Sublevel: Blue or white square

Subject Areas

When you create an analysis, you first select a subject area, which contains columns related to a specific business object or area. You then open folders within the subject area to find the columns to include in your analysis.

Folders

Each subject area has one fact folder and a number of dimension folders. Folders can have subfolders.

• Fact folders:

  • Contain fact columns.

  • Are usually at the bottom of the list of folders and are usually named after the subject area.

• Dimension folders:

  • Contain attribute and hierarchical columns.

  • Are joined to the fact folder within a subject area.

    For example, if your analysis has the Currency attribute from a dimension folder, you see currencies in the results. If you also add the Total fact, then your analysis includes only records with both a currency and a total amount. The more columns you add, the smaller the query set for your analysis.

  • Can be common folders or common dimensions that appear in more than one subject area.

    Note: If your analysis has columns from multiple subject areas, then you: