Referent Resolution and Discourse Relations

• Referent resolution concerns the ways in which we use language to refer to entities that have already been mentioned (or which can be inferred), and the strategies we use to figure out what is being referred to.
  • Methods of referring to previously mentioned entities in discourse
  • What factors govern the choice among these methods?
  • How can we encode this information in NLP systems?
• Discourse relations are the explicit and implied relations between sentences, clauses, and other parts of a discourse that make it possible to interpret it coherently.
  • What kinds of relations are most common between parts of a discourse?
  • What kind of overall structure does discourse display?
  • How can NLP systems detect and exploit these relations and structures?

(2)

(back to syntactic ambiguity resolution)

(return to syllabus)