Published at: 2026-09-17
Semantic retrieval index admin guide
Create a Semantic Retrieval Index
In AI Platform Management > Semantic Retrieval Index, you can view and create knowledge indexes. Built indexes can be called by AI assistants, prompt templates, and other AI features.
Click New in the upper right and select an index type.
Index types fall into two categories:
- Local Document: Upload files directly when enterprise knowledge is maintained as PPT, Excel, or similar files.
- Business Object: Use system business data for Q&A, or import data from system objects when knowledge already lives there.
Local Document
- Configure basic info, including Name and Description.
- Upload the files to index and set segmentation for each file.
- Different file types have different segmentation options.
- After segmentation, preview the chunk content.
Business Object
Business object indexing has three steps: Basic Settings, Data Source Settings, and Index Field Settings.
- Basic Settings: Configure name and description, similar to local document upload.
- Data Source Settings: Select the business data range.
- Index Field Settings: Select a title field and a knowledge content field, then configure segmentation for the content field.
Different field types have different segmentation rules. The title field and chunk field values are added to each chunk.
- After segmentation, preview the chunk content.
Semantic segmentation strategies
Select a title field for knowledge
When segmenting an object, select a title field. During retrieval, the system searches the title field in addition to semantic and full-text recall. It treats the segments that match the title field as a separate recall path, then reranks all paths to produce the most complete answer.
Tip: Choose a field with business meaning that users query most often.
Add a chunk field
Segmentation can include a chunk field. The chunk field value is added to each segment as semantic reinforcement.
Tip: Choose a field with business meaning that users query most often.
Document hierarchy in segmentation
For document-type indexes, the system adds the document hierarchy of each segment to every chunk (for documents with a hierarchy). Hierarchy adds relationships and topic context to segments, which improves recall accuracy and quality. If a question matches a hierarchy node, all segments under that node are recalled to generate the answer, which improves recall and precision.