Hybrid Search Autodrop Results

Dynamically filter out results that show a sharp decrease in relevance scores to increase the precision of search results for your prompts and RAG use cases. Data 360 hybrid search autodrop is a method by which you can filter out irrelevant search results that can mislead upstream applications like prompt builder using those results.

When you initiate a search query, the hybrid search fusion ranking model derives an optimal ranking score for each search result it retrieves. The hybrid search autodrop option allows you to drop the results from one or more score cutoff points where the scores drop significantly. You can consider results beyond this cutoff point as less relevant to the query.

Autodrop is available only with hybrid search. Since hybrid search combines both vector search and keyword search, you can use query API to configure autodrop for the results of each search type in hybrid search, including vector search results, keyword search results, and the fused hybrid search results. Hybrid search analyzes the search result scores and filters irrelevant results. Hybrid search autodrop is disabled by default. Enable autodrop using Data cloud Query API.

• Hybrid Search Query Expressions for Autodrop
  Use Data 360 Query API to configure autodrop for hybrid search results.