Skip to content

Natural language search #119

@thibaudcolas

Description

@thibaudcolas

Tentative item, highly dependent on evolution of vector indexing / search capabilities across database backends. Improvement to Wagtail’s search interface to support natural language queries.

Intended outcome

Depending on underlying search backend capabilities, this would either be:

  • Vector search, with semantic understanding of queries and content.
  • Or RAG (retrieval-augmented generation) approach, combining traditional search with LLMs to interpret queries and content.
  • Or (TBC) conversational search interface, with follow-up questions and clarifications.

More information

Metadata

Metadata

Assignees

No one assigned

    Labels

    scaleHigh-traffic, multi-site, large content sets, reliability at scale.searchSearch UX, indexing: query, ranking, Elasticsearch and database search integrations.

    Type

    No type

    Projects

    Status

    Future

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions