Create an Intelligent Context Configuration

Enable Agentforce

Create a Configuration

1. In Data 360 home, go to the Process Content tab.
2. Select Intelligent Context in the left-hand navigation and click New Configuration.
3. In the New Search Configuration dialog, choose a data space and name for the search configuration, and click Save.
4. From the Intelligent Context page, select your new project.
5. Click Upload Files or Choose different source to select a DMO.
6. Select a file or a DMO to review its content in the Configuration Preview.

Configure with AI Smart Defaults

1. In the agent chat, click Set up my configuration using smart defaults.
2. Review the configuration summary, and choose to accept or modify it.
   1. If you accept the configuration, continue to Test and Evaluate a Configuration.
   2. If you want to modify the configuration, continue to Modify an Intelligent Context Configuration.

Modify an Intelligent Context Configuration

1. Click the Edit Configuration tab.
2. To change the search type, click Select Search Index Type, and choose vector or hybrid search.

   Vector	Index content based on semantic meaning.
   Hybrid	Index content based on semantic meaning and keywords.

3. To change how your content is parsed, click Set Parsing Options.
   You can't select both LLM-based parsing and LLM-based Visual Data Preprocessing options. Review LLM-based parsing and preprocessing for more information about these options.
4. Choose one of these options.

   Default Parser	Extract text using built-in platform settings.
   LLM-based Parser	Extract all text, images, and other visual elements using a large language model (LLM).
   Docling Parser	Extract text from images, tables, and document layouts using the Docling parser.

5. To change how your content is preprocessed for chunking multimodal elements, such as images and tables, click Set Preprocessing Options.

   Preprocessing extracts context from images and extracts data from tables while maintaining context and relationships.

6. From the Preprocessing dropdown list, choose one of the following:

   No preprocessing	Default setting.
   LLM-based preprocessing	Use the default prompt as is, or edit it for your specific use case.

7. To extract text from tables, forms, and image captions, turn on Image Processing.
8. Click Select an Embedding Model, and choose an embedding model from the list.
9. To change how your content is chunked, click Set Data Chunking Rules, and choose a chunking strategy.
10. Adjust the token settings as needed.

    Chunking Strategy	Description
    Section-Aware Chunking	Uses title and heading elements to chunk documents. Additional settings: Max Tokens: Combine adjacent small sections into larger chunks by adjusting the token size. Overlap Tokens: Use only if you create multiple chunks from a single section. Specify the number of tokens to add from the end of each chunk to the beginning of the next chunk.
    Passage Extraction	Uses the semantic meaning inherent in HTML tags to chunk a document into passages. Additional settings: Max Tokens: Combine adjacent small sections into larger chunks by adjusting the token size.

11. Turn on Knowledge Harmonization to render your content in a consistent viewing experience without merging the source content.
12. To add and map metadata fields, such as subject and file type, for local file uploads, select SI Optimization. Use these fields to apply prepend values to your generated document chunks, or to set pre-filter conditions that refine your search results.
13. Select the metadata fields to prepend to the generated document chunks.
14. Add pre-filter conditions to refine query results based on your metadata field values.
15. Click Apply Changes.

    When you click the button, Data 360 automatically generates and displays a new set of content chunks in the Configuration Preview pane.

16. Continue to test and evaluate your configuration.

Test and Evaluate a Configuration

1. In the agent chat pane, select a suggested question or enter your own query in the chat box.
2. To open the source file in the preview pane, click a generated chunk. You can also search or filter the list of chunks.
3. To compare the generated chunks directly with the source document, click Compare with Source.
4. To make further changes to your configuration, follow the steps in Modify an Intelligent Context Configuration.
5. Continue to publish your search index configuration.

Publish the Search Index Configuration

Before you can publish your configuration, you need to set up an unstructured data connection to ingest your source files. Set up an unstructured data connector to ingest your source files into an unstructured data lake object (UDLO) in Data 360. See Unstructured Data File Formats and Connectors for a list of connectors.

Note If you've selected a DMO to create the configuration, you don't need to set up an unstructured data connector.

1. From Intelligent Context, click Publish.
2. In the Select an Unstructured Data Model Object dialog, select a data model object to apply the configuration to.
3. Click Next.
4. In the Create Search Index dialog, confirm or change the index name, and click Publish.
5. Navigate to the Content Indexing or Search Index pages to review the results.
6. To manage and review your harmonized objects in Data 360, go to the Knowledge Harmonization page under Process Content.