Algorithms

Recommend For Ranking

Empirical algorithm recommendation for a specific ranking.

Combines the rule-based registry with backtesting on the ranking's actual vote history. If the ranking has fewer than 100 votes, falls back to rule-based. If it has more, runs paired-bootstrap log-loss comparison across {elo, glicko2, bradley_terry, openskill, elo_mmr} and returns the empirically-best algorithm with statistical evidence.

This is the SOTA-aligned upgrade over /algorithms/recommend.

get/api/v1/algorithms/recommend-for-ranking/{ranking_id}

Request

  • Base URL: https://app.aimrank.io (the document declares no server; this is the origin it was published from)
  • URL: https://app.aimrank.io/api/v1/algorithms/recommend-for-ranking/{ranking_id}
  • Auth: none declared

Path parameters

ranking_idstring required

Query parameters

domainstring nullable

Domain hint (sports_1v1, products, etc.)

has_teamsboolean

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 1 revision of this API.