Set Page Rank
Create a step-retrieval strategy using page rank.
Scenario
Section titled “Scenario”In an AI-powered chat, you can configure a chat assistant or an agent to respond using knowledge retrieved from multiple specified knowledge bases, provided they use the same embedding model. In situations where you prefer information from certain knowledge base(s) to take precedence or to be retrieved first, you can use the page rank feature to increase the ranking of chunks from those knowledge bases.
For example, if you have configured a chat assistant to draw from two knowledge bases — knowledge base A for 2024 news and knowledge base B for 2023 news — but wish to prioritize news from 2024, this feature is particularly useful.
Configuration
Section titled “Configuration”On the Configuration page of your knowledge base, drag the slider under Page rank to set the page rank value. You can also input the intended value directly in the field next to the slider.
Mechanism
Section titled “Mechanism”If you configure a chat assistant’s similarity threshold to 0.2, only chunks with a hybrid score greater than 0.2 × 100 = 20 will be retrieved and sent to the model for content generation. This initial filtering step narrows down relevant information.
If you assign a page rank of 1 to knowledge base A (2024 news) and 0 to knowledge base B (2023 news), the final hybrid scores are adjusted accordingly:
- A chunk from knowledge base A with an initial score of 50 receives a boost of 1 × 100 = 100 points, resulting in a final score of 150.
- A chunk from knowledge base B with an initial score of 50 receives no boost, maintaining a final score of 50.
Chunks from knowledge base A will always precede chunks from knowledge base B in the retrieval results.