Logical dimension table
A dimension is an object for statistical analysis. It is typically a real-world entity. Dataphin follows Ralph Kimball's Dimensional Modeling theory. You create dimensions to define business entities, also known as master data, and ensure their uniqueness. Dimensions and their combinations also define the statistic granularity for derived metrics. For example, when you analyze a transaction, you can use dimensions such as buyer, seller, product, and time to describe the context of the transaction. This topic describes logical dimension tables.
Prerequisites
You must have created a business entity. For more information, see Create and manage business entities.
Access the configuration page for logical dimension tables
On the Dataphin home page, click R&D in the top menu bar.
Follow the steps in the figure to open the Logical Dimension Table configuration page.
Configure a logical dimension table
Dataphin lets you create several types of logical dimension tables.
Logical table type | Description |
A normal logical dimension table describes an entity and its attributes. For example, a normal logical dimension table for members contains data such as member name, member ID, and member email. | |
A hierarchical logical dimension table represents dimensions that have a natural parent-child structure. For example, a hierarchical logical dimension table for regions contains data such as country, province, city, and district. | |
An enumeration logical dimension table standardizes a list of values for an enumeration dimension. For example, a lookup table with a key-value structure can be used for gender. | |
A virtual logical dimension table standardizes dimensions that are not associated with a specific business entity and do not have a definable data scope. Examples include visitors and URLs. |