Recombee alternatives

Recombee is a recommendation API. You push a catalogue of items and a stream of interactions, then call endpoints such as Recommend Items to User, Recommend Items to Item and Search Items, and get back a ranked list. It is one of the closest products to NeuronSearchLab in shape: API-first, catalogue-and-events in, ranked slate out, no front-end widget required.

Its distinguishing piece is ReQL, a query language for filtering and boosting at request time, and the scenario concept, which lets each placement on your site carry its own configuration. Model behaviour is largely chosen for you: Recombee documents automatic algorithm selection rather than a catalogue of architectures you pick between.

Teams typically reach for Recombee when they want personalised recommendations running quickly, on a managed service, without operating any modelling infrastructure themselves.

Read it from the source: Recombee documentation.

A single recommendation model gives you one fixed way of ranking. NeuronSearchLab trains and compares several architectures on your own data, keeps the winner behind one unchanging API, and gives you evaluation, experimentation, explainability and editorial control over whatever is serving.

Against Recombee specifically, the differences worth checking are model selection, bring-your-own embeddings, per-model evaluation and explainability - set out row by row on the NeuronSearchLab vs Recombee page, with each of Recombee’s cells sourced from its documentation.

If you want independent evidence rather than a vendor’s word, the NSL Recommender Leaderboard measures complete recommender systems on ten public datasets with the full methodology, the caveats and the downloadable results published.