C3 AI Documentation Home

Manipulate ERD Views with Object Model

Data Fusion is in Beta. Please contact your C3 AI representative to enable this feature.

After integrating sources and mapping them to C3 AI Entity Types, you can use the Object Model tab to review entities and their relationships through primary and foreign keys. While C3 AI Console allows you to view entities in a table view, Data Fusion's Object Model tab shows an Entity Relationship Diagram (ERD) to surface the overall structure of data within your integration.

The Object Model tab follows standard ERD conventions with a few notable exceptions:

  • Entity color coding — An entity's border color indicates the package it belongs to by default. Modify color options in Diagram Settings.
  • Mixin listings — Find an entity's mixins underneath the entity name.
  • Mandatory field markers — You can identify a field as mandatory when it's preceded by an exclamation point (!). This notation follows standard C3 AI Type syntax.
  • Keys — Identify primary and foreign keys by key symbols. The Legend in the bottom left corner displays examples of primary, foreign, and multiple foreign keys.
  • Calculated fields and member functions — These parts of entities are displayed below all stored fields. The legend explains how to differentiate between read calcs, stored calcs, and member functions. You can find more about mapping data to fields in Map and Transform Source Data
  • Selectable entities — Select an entity to review its properties in more detail and preview data.

Entities and their relationships

The Object Model tab enables you to visualize entities, allowing you to manipulate, filter, and simplify views to show what you value most in your application.

Entities within Object Model represent entity and Canonical Types in your application. You can use the ERD to review these types and understand their sources.

Legend

Use the Legend to review color indications, subtitle selections, and explanations of icons. Hide the Legend by selecting the icon of an eye with a slash through it.

Search

Use the search bar to navigate to an entity. As you type, this feature lists possible nodes for easier searching.

Diagram Settings

Use these options to simplify your diagram. Manipulate the following in Diagram Settings:

  • Color coding — Change colors to indicate how entities receive data.
    • Populated by App indicates that the data resides within the app. An Alerts entity is a good example of data that is created and maintained within the app.
    • Needs to be integrated indicates that the you use a pipeline to feed the data into this entity from a database, a streaming source, a CSV upload, or a file system.
  • Tags — Manipulate tags to show details such as whether an external data source is hosting the data (which is true of any SQL-based data source), which entities remix other entities, and whether the data belongs to a Key-Value store.
  • Details — Choose whether entities list their fields, calculated fields, or methods.

Toggle views

By the search bar, select the dropdown to choose between a Detailed or Minimalist view. This option allows you to remove all details of what the entities show to focus exclusively on how entities relate to each other.

Closeup of the Search Bar, the Filter selection, and the Views selection

Any time you move entities in your diagram, Save and Reset buttons appear to the right of the views dropdown. These buttons allow you to save or revert all changes you made to the selected view, including any changes to Filters or Diagram Settings.

Saved changes apply to the entire application. If you and a colleague are working on the same app, your changes appear in their Object Model view and vice versa.

Filter entities

The Filter option in the Object Model tab helps you control which packages are displayed in the workspace. The Packages filter lets you customize the Object Model view to display selected packages, keeping your workspace focused, organized, and easier to navigate.

In C3 AI Studio, a package is a logical grouping of related entity types, mixins, relationships, and services, for example, genAiSearch, observability, or uiComponentLibrary.

Filtering allows you to focus only on the specific parts of the data model that are relevant to your current work.

The list displays all packages in the root package of your application and its dependencies. Each package represents a collection of entities.

Some packages may be excluded from appearing in the Object Model view based on your system configuration. Administrators can manage this behavior through the EntityRelationshipDiagramPackageConfig#excluded setting.

Examples include:

  • genAiSearch – Package for AI-driven search features.

  • observability – Entities for monitoring and telemetry.

  • uiComponentLibrary – UI and component-related models.

To include or exclude a package from the Object Model view

Users with the C3.AppAdmin role can control which packages appear in the Object Model view using the EntityRelationshipDiagramPackageConfig#excluded setting.

Follow these steps to view or update the configuration:

  1. Open the C3 Console in your environment.

  2. Retrieve the current configuration for a specific package using the following command:

TypeScript
EntityRelationshipDiagramPackageConfig.make({
    name: 'selfServiceDI'
    }).getConfig()

    * Replace 'selfServiceDI' with the name of the package you want to check. 
    * The output displays the current configuration, including the excluded value. 
    * If excluded is set to true, the package is hidden from the Object Model diagram.
  1. Update the configuration to include or exclude the package
TypeScript
EntityRelationshipDiagramPackageConfig.make({ 
     name: 'selfServiceDI', 
     exclude: false/true 
 }).setConfig()

    * Set exclude: false to include the package in the diagram. 
    * Set exclude: true to hide the package from the diagram.

  1. Refresh the Object Model view to apply the change.

  2. Select All checkbox

    • By default, Select All is enabled, meaning all packages are displayed in the Object Model view.
    • You can deselect individual packages to narrow your view to only the ones you need.

  3. Clear button

    • Clears all selections, hiding all packages temporarily from the Object Model visualization.

  4. Apply button

    • After selecting or deselecting packages, select Apply to update the Object Model canvas to reflect your filtered selection.
    • Only entities belonging to the selected packages will appear in the view.

  5. Search bar (above the filter)

    • Use the search bar to find a specific entity or node by name within the currently visible packages.

Filtering packages helps:

  • Simplify the Object Model view by hiding unrelated packages.
  • Improve performance when dealing with large or complex data models.
  • Focus analysis on the entities you’re actively developing or integrating.

Add and update entities

Your application must be in development or trial mode to edit entities. You cannot edit production or test applications from Data Fusion.

Data Fusion offers a selection of entity templates to begin; you can also select Create an Entity from Scratch if none of the templates suit your needs. For more information on the entities templates, see the documentation on the following Types:

Create an entity from a file upload

The Create entity from file upload feature allows you to create a new entity in your application model by uploading structured files (such as CSV) directly into C3 AI application object model. This process is useful when you want to bring data from external flat files into the C3 AI platform for modeling or validation.

Select files

  1. Navigate to Data Fusion > Object Model.

  2. Select + and select Create entity from file upload. The interface opens to the Select files step.

  3. Select Browse to choose one or more files to upload.

    • Supported formats: .csv

    • Total file size limit: 2000 MB

  4. Once a file is selected, it appears in the Files list below, showing:

    • File name – The uploaded file name.

    • Status – Indicates whether the file is awaiting upload or processed.

    • Size – Displays the file size.

    • Warnings – Displays any schema or data issues detected during preview.

Uploads do not begin until you confirm the schema in the next step.

You can also enable Data lake sync if you want to automatically synchronize uploaded files to the configured data lake storage. This option is off by default.

Configure schema and preview

After selecting your files, proceed to the Configure schema and preview step. The Configure schema and preview step allows you to review and refine how the system interprets your uploaded data before creating an entity in the C3 AI Object Model.

This step ensures that your data structure (columns, data types, and schema) aligns with the application model and is ready for successful ingestion.

When a file (CSV) is uploaded, C3 AI Studio automatically detects and infers its schema based on the file’s content. The Configure schema and preview step lets you verify and edit that schema before any data is processed or committed.

Here, you can:

  • Inspect the first 20 rows of your data.

  • Review automatically inferred column names and data types.

  • Make adjustments to align with your desired schema before ingestion.

Preview entity with schema

The Preview Entity with Schema tab allows you to verify how your uploaded file appears after the configured schema has been applied.

This step helps ensure that data columns, types, and values are correctly interpreted before creating an entity in the C3 AI Object Model.

  1. Select a file to preview

    a. From the file list on the left, choose the file you uploaded (for example, melanoma_Immunotherapy_trial_data_1.csv).

    b. C3 AI Studio applies the schema you configured in the previous step to the file.

  2. Preview data with schema applied

    a. The table on the right displays the first 20 rows of your dataset after the schema has been applied.

    b. You can confirm that:

    • Column headers are correctly named and aligned with their aliases.

    • Data types are correctly interpreted (for example, numbers are shown as numeric values, not text).

    • The data appears consistent and formatted.

  3. Inspect column layout and sample values

    a. Each column header shows its corresponding alias and data type.

    b. You can visually check if all columns were recognized correctly and whether any data appears truncated, misaligned, or incorrectly parsed.

  4. Refine if needed

    a. If you spot discrepancies, such as mismatched column names, wrong data types, or missing fields, click Back to return to the Entity Schema tab and adjust the schema before proceeding.

  5. Proceed once validated

    a. When you’re satisfied with how the data looks, click Next to move to the final step: Configure entity properties.

This preview helps ensure that:

  • The schema you defined correctly matches your data’s structure.

  • There are no parsing errors or unexpected values.

  • The resulting entity will represent your data accurately once created in the Object Model.

In short, The Preview Entity with Schema tab is your quality checkpoint, it lets you visually confirm that the uploaded file aligns with your defined schema before finalizing the entity creation.

Configure entity properties

In this final step, you define key details for the new entity that will be created from your uploaded file. This ensures your uploaded dataset is registered within the C3 AI Object Model so it can be used for analysis, integration, and visualization.

  1. Enter an entity name

    a. Provide a unique name for the new entity.

    b. The name should follow your organization’s naming standards (for example, use camelCase or PascalCase).

    Example: MelanomaImmunotherapyTrialData1.

    Avoid using spaces or special characters.

The entity name becomes the identifier that other users and applications in C3 AI Studio will use to reference this dataset.

  1. Add a description

    a. In the Description field, describe the content and purpose of this dataset.

    b. Include key context such as data source, scope, or what the data represents.

    Example: Contains anonymized patient data from melanoma immunotherapy clinical trials, including age, genetic markers, treatment response, and survival metrics.

    Adding a description helps improve data discoverability and collaboration across teams.

  2. Review and upload

    a. Once the name and description are provided, select Upload (bottom right).

    b. The file gets processed and a new entity gets be created in your application model.

    c. You can monitor upload progress and review the new entity once it appears under the Object Model tab.

  3. Optional: click Back

    a. If you need to make schema or mapping adjustments, you can select Back to return to the Configure schema and preview step.

After the entity upload completes:

  • The dataset becomes accessible as a structured entity in the Object Model.

  • You can view its schema, explore records, and use it in downstream workflows such as Data Integration, Feature Engineering, or Data Validation.

Edit an existing entity

To edit an existing entity, right-click the entity and select View or Modify Code.

After editing the new or existing entity, select Save to update the entity within your application. These changes persist until someone upgrades the application. To save at the package level, you must then push your changes to GitHub. For more information on pushing changes to GitHub, see Version Control with GitHub.

You can only update entities within the root package of your application. If you modify entities from dependency packages, the edits save to a new remixed entity in your root package.

Download your ERD

The Object Model tab offers an option to download your ERDs for sharing information and data flows with stakeholders. Before you select download from the upper left corner, consider using the Filter dropdown menu to deselect any unnecessary packages.

Download the diagram to an SVG file.

See also

Was this page helpful?