Skip to main content

Semantic Model Series: 5. Presentation Layer

Image
By

Oracle Analytics Bootcamp

In Oracle Analytics Semantic Modeler, the Presentation Layer is the final layer in the data modeling flow — it's what end users interact with when building analyses and dashboards.

The Presentation Layer provides a business-friendly view of your data model. It organizes and simplifies logical objects (tables, columns, hierarchies), and allows you to control the visibility, naming, and grouping of elements within subject areas.

Semantic model

Components of the Presentation Layer

  • Subject Areas

    • A subject area is a collection of related tables and measures — such as “Sales,” “Finance,” or “Inventory.”
    • Each subject area typically maps to a single Business Model from the logical layer.

  • Presentation Tables

    • Represent business entities like “Customer,” “Product,” or “Revenue Metrics.”
    • These can be renamed to more user-friendly labels compared to their logical counterparts.

  • Presentation Columns

    • Derived from logical columns in the Business Model and Mapping (BMM) layer.
    • You can rename, reorder, or hide columns as needed.

  • Hierarchies and Folders

    • Hierarchies (e.g., time or geography) allow for drill-down functionality.
    • Columns can be organized into folders or nested groups for better clarity.

  • Column and Table Permissions

    • Control which users or roles can see or interact with specific elements.
    • Essential for data security and for tailoring the experience to different audiences.

Common Tasks in the Presentation Layer

  • Renaming tables and columns to reflect business terminology.
  • Hiding technical or irrelevant columns from end users.
  • Organizing columns into folders like “Sales Metrics” or “Customer Details.”
  • Controlling display order, tooltips, and descriptions.
  • Applying formatting to columns (e.g., date or currency formats).
  • Enabling drill-down paths through defined hierarchies.

Creating a New Subject Area

The Presentation Layer can be accessed via the fourth tab in the main Semantic Modeler navigation. When creating a new model from scratch, this layer starts off empty.

Presentation Layer - empty

All content in the Presentation Layer is organized by Subject Areas.

A Subject Area is the top-level container within the Presentation Layer. It represents a logical grouping of data that users interact with when building dashboards, reports, or ad-hoc queries. Think of it as the user-facing lens into your data model.

A common best practice is to create one Business Model in the logical layer and map multiple Subject Areas to it. This enables users to pull data from different business domains and combine it in a single report — as long as shared dimensions exist across the subject areas.

To create a new Subject Area, click the plus icon next to the Search bar and select Create Subject Area from the dropdown menu.

Create Subject Area - menu

In the Create Subject Area dialog, provide a meaningful name for the new subject area.

Create Subject Area - dialog

Once created, the Subject Area page opens, which includes several definition tabs:

  • General: Define the name, description, and link the subject area to its associated business model.
  • Tables: Add and manage presentation tables and columns, including hierarchies and folders.
  • Permissions: Set visibility rules for different user roles or groups.
  • Localisation: Provide multilingual labels and descriptions for internationalized deployments.

Adding Tables to a Subject Area

In the Tables tab of the Subject Area, the task is to bring in tables and columns from the logical model. There are multiple ways to do this, but since our logical model already includes meaningful names and structure, we’ll use a shortcut.

Switch to the Logical Layer tab in the main navigation and expand the existing business model to display all tables. Select the tables you want to import.

Expand business model and select tables

With the Subject Area open in the Tables tab, drag the selected tables from the Logical Layer pane and drop them onto the Tables canvas.

Drag and drop Logical table into Subject area

Once added, you can reorganize the table order for a cleaner view:

Reorganize tables in Subject area

You can also drill into any presentation table for further customization by double-clicking its name. This opens the Presentation Table page, where the following options are available:

  • General
  • Columns: Rename columns or replace the associated logical column.
  • Hierarchies: Create or edit drill hierarchies.
  • Permissions: Set access control at the column level.
  • Localisations: Add translations or locale-specific labels.
  • Data Filters: Apply filters to restrict data exposure at the presentation level.

No further changes are needed for now, so we can move on to the final step — deploying the model.

Deploying the Semantic Model

Before deploying, it’s important to validate the model by running a Consistency Check.

Consistency Check ensures the semantic model is logically sound, fully defined, and free of errors before being published.

You should run a Consistency Check:

  • After changes to logical models, joins, hierarchies, or subject areas.
  • Before deploying the model to production.
  • When troubleshooting query or report issues.

Click the Consistency Check icon in the top toolbar:

Run consistency check

If there are no errors (you may still see warnings), a report summary is displayed:

Consistency check report

Once validated, deploy the model by clicking the page menu (three-dot icon in the top right) and selecting Deploy.

Deploy model

In the Deploy Semantic Model dialog, click Deploy to finalize.

Deploy Semantic Model dialog

If everything is successful, a confirmation appears in the bottom-right corner:

Deploy Successful

Testing the Model in Oracle Analytics

Now that the model is deployed, we can test it in Oracle Analytics.

Start by creating a new Data Visualization Workbook:

Create a new Workbook

In the Add Data dialog, search for the newly created Subject Area and click Add to Workbook.

Add Data to Workbook

Select a few columns, for example:

  • D Customer.Customer Segment
  • D Orders.Channel
  • F Revenue.# of Orders
  • F Revenue.# of Products
  • F Revenue.Revenue

Create Best Visualisation

Right-click on any of the selected columns and choose Create Best Visualization.

If the model was set up correctly, Oracle Analytics will automatically generate a meaningful chart — in this case, a scatter chart.

Create Best Visualisation displayed

This concludes our walkthrough of the Presentation Layer in Oracle Analytics Semantic Modeler. In this session, we:

  • Created a system connection and initialized the Semantic Modeler.
  • Built the Physical and Logical Layers with facts and dimensions.
  • Designed and deployed a new Subject Area in the Presentation Layer.
  • Successfully tested the model in Oracle Analytics.

In the next session, we'll dive deeper into some advanced topics that are essential for building powerful and flexible semantic models.