Skip to content

About

Displays related nodes using AI embeddings and vector search.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

AI Related Content

Displays related nodes using AI embeddings and a vector search.

Installation

Popularity (trending) blending

Related results can be re-ranked with recent page views so trending items among the related candidates rise. Unrelated popular content is never added.

  • A small JavaScript beacon on full node pages posts one view per node per browser session to ai-related-content/hit, so views served from the page cache still count. Bots, users with "administer nodes", and IPs over the hourly limit are ignored.
  • Views are stored as daily counts in ai_related_content_views and pruned on cron after the configured window.
  • Ranking: blended = (1 - w) * similarity + w * trending, both normalised to 0..1 within the candidate pool. Trending uses exponential decay (configurable half-life) and log1p damping.
  • While blending is on, results are cached for the shorter popularity cache lifetime instead of one week.

Configure under Popularity (trending) at admin/config/ai/ai-related-content. Blending is off by default: switch on "Blend popularity into related content ranking" to use it. View tracking is a separate setting, so history can build up before blending is switched on. Turning blending off restores pure similarity ranking. The test tab shows similarity, trending views and blended score.

Note: the Views block narrows results to the chosen node IDs, then applies the View's own sort. Blending changes which items appear there; the legacy block also keeps the blended order.

Issues

Bugs and feature requests should be reported in the Issue Queue.

Current Maintainer

Justin Keiser

Credits

License

This project is GPL v2 software. See the LICENSE.txt file in this directory for complete text.

About

Displays related nodes using AI embeddings and vector search.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages