Reference in Text

Machine learning of anaphora resolution and information structure

Event date: November 8, 2012 2:30 PM


The workshop was organized by Heike Zinsmeister as an activity of CLARIN-D and the Collaborate Research Centre "Incremental specification in context" (SFB 732).

Topics (program with abstracts):

  • Anaphora and coreference resolution -- including the resolution of
    abstract anaphora
  • Machine learning of information status categories and how they
    interact with anaphora resolution
  • Active learning and crowd-sourcing methods for creating the
    training data (e.g., Amazon Mechanical Turk)
  • Fine-grained quantitative error analysis and evaluation of
    corefence resolution systems that are used on large data sets
    (e.g. the CONLL coreference data sets)
  • Domain adaptation in anaphora and coreference resolution