RuleTreeRank (RTR) is an interpretable Learning-to-Rank framework that models ranking as a two-stage process: it first groups items with a shallow rule tree, then refines the score locally through instance-based comparisons.
learning-to-rank explainable-ai interpretable-machine-learning rule-based-explanations instance-based-reasoning ruletreerank mixruletreerank
-
Updated
Sep 20, 2026 - Python