milia.recommend#
Retrieval, baselines, case-based action recommendation and ranking.
Modules
Case-based recommendation: suggest the action that fixed the most similar past cases. |
|
Baselines a retrieval method has to beat. |
|
Diagnostics for whether entity-level retrieval makes sense. |
|
Equity of a selection or a burden: how evenly it falls on people or places, and selection under a bound. |
|
Learning to rank: order items within each list by a score learned from their features. |
|
Ranking cases, groups and models, with uncertainty on the ranks. |
|
Content-based retrieval: predict an entity's outcome from its nearest labelled neighbours. |
|
Uplift: the change in an outcome that an action causes, estimated for each case from its features. |