What it does
The engine has a co-purchase recommender and a search helper that boosts results by co-occurrence with a seed product. No route, command or tool calls either, so a running server cannot be asked for recommendations today.
How it works
The recommender scans the order entities. Each order is expected to carry an items array of product ids. For every pair of products that appear in one order it counts a co-occurrence; for a seed product it returns the products most often bought with it, highest count first, ties broken by id. A per-user variant gathers the products a user has bought, sums co-occurrence across all orders and returns the top products the user has not bought yet.
A search helper takes a normal product query and a seed product, normalises the co-occurrence counts to the range 0 to 1 and adds a weighted share to each hit score before re-sorting.
Where you reach it
- Library
Recommender::recommend, Recommender::recommend_for_user and ProductSearch::with_recommendations (Rust engine API; not exposed by the server)
Limits
- Not reachable through any route, command or tool of the private-preview server.
- Every call rescans all order entities and builds the pair counts in memory, so cost grows with orders and with items per order; the module suggests pre-building the matrix offline at large scale.
- Orders must use the items array of product id strings; other shapes yield no signal.
Evidence
- Tests3 library tests cover top-k co-purchase results, an empty store, and exclusion of products a user has already bought
Test coverage is listed as present in the engine repository at this edition. This page does not report test results.
KynetraDB is in private preview
This is not available in the preview. Request a preview to discuss the current scope.