Item Ranking Task

AKA: Scoring.
Context:
- Input: an Item Set; a Ranking Function.
- output: a Ranked Output (an Ordinal Value List)
- Measure(s): a Ranking Task Performance Measure.
- It can range from being a Deterministic Ranking Task to being a Predictive Ranking Task (e.g. ordinal prediction)
- It can be solved by an Item Ranking System (that applies an item ranking algorithm).
Example(s):
- Predict whether a competitor will come in 1st place, 2nd place, 3rd place, or other.
- Predict whether a person will rank a movie as excellent, good, okay, boring, bad.
- an Entity Ranking Task, such as graph node ranking or sentence ranking.
- a Prioritization Task that ranks which tasks are more important given their interdependence.
- an Information Retrieval Task.
- a Relevance Ranking Task.
- …
Counter-Example(s):
- a Classification Task.
- an Estimation Task.
See: Quantification Task, Decisioning Task.

References

http://en.wikipedia.org/wiki/Enumeration
- In mathematics and theoretical computer science, the broadest and most abstract definition of an enumeration of a set is an exact listing of all of its elements (perhaps with repetition). The restrictions imposed on the type of list used depend on the branch of mathematics and the context in which one is working. In more specific settings, this notion of enumeration encompasses the two different types of listing: one where there is a natural ordering and one where the ordering is more nebulous. These two different kinds of enumerations correspond to a procedure for listing all members of the set in some definite sequence, or a count of objects of a specified kind, respectively. While the two kinds of enumeration often overlap in most natural situations, they can assume very different meanings in certain contexts.

(Li & Link 2006) ⇒ Ling Li, and Hsuan-Tien Lin. (2006). “Ordinal Regression by Extended Binary Classification.” In: Advances in Neural Information Processing Systems 19 (NIPS 2006).
- We work on a type of supervised learning problems called ranking or ordinal regression, where examples are labeled by an ordinal scale called the rank. For instance, the rating that a customer gives on a movie might be one of do-not-bother, only-if-you-must, good, very-good, and run-to-see. The ratings have a natural order, which distinguishes ordinal regression from general multiclass classification.