Word Sense Clustering Task

A Word-Sense Clustering Task is a Word-level Analysis Task is a Coreference Resolution Task (requires the grouping of all Words in a Corpus that refer to the same Concept).

• AKA: Word Sense Discrimination Task, Word Sense Coreference Resolution Task.
• Context:
  • Input: a Corpus.
    • Optionally: an Entity Type Definition.
  • output: a Word-Sense Inventory.
    • Optionally: a mapping to the Word Mentions in the Corpus.
  • It can be solved by a Word Sense Clustering System (that implements a Word Sense Clustering algorithm).
  • It can be solved by a Word-Sense Discrimination System (that implements a Word-Sense Discrimination algorithm).
• See: Word Sense Disambiguation Task, Homonym Relation, Entity Mention Coreference Resolution Task, SemEval Task.

---

---

References

2005

• Rada and Pedersen (2005). “[Advances in WSD.]] Tutorial
  • Word sense discrimination is the problem of dividing the usages of a word into different meanings, without regard to any particular existing sense inventory.

1998

• (Schütze, 1998) ⇒ Hinrich Schütze. (1998). “Automatic Word Sense Discrimination.” In: Computational Linguistics 24(1).

---