Skip to content

Latest commit

 

History

History
59 lines (37 loc) · 2.14 KB

negation-annotation-task.md

File metadata and controls

59 lines (37 loc) · 2.14 KB
Raw

Negation Annotation in Dutch dialogue

Metadata

  • Status: Completed
  • Type: Specific
  • Work Package: WP3
  • Coordinators: Tom Sweers (Radboud), Iris Hendrickx (supervisor, Radboud)
  • Coordinator in CLARIAH: Maarten van Gompel (HuC)
  • Participating Institutes: Radboud University Nijmegen
  • End-users: (Who is involved as end-user for this use-case? Try to mention name, institute, role/responsibility)
  • Developers: Tom Sweers (Radboud), Maarten van Gompel (HuC)
  • Interest Groups: Text
  • Task IDs: T062 (FLAT), T108 (FoLiA)

Description

What is the research about?

Citation from research plan by Tom Sweers:

The aim of this research is to investigate negation detection models in Dutch dialogue. A small corpus will be annotated, multiple models will be experimented with.

The context of this research is the BLISS Project (Waterschoot et al. 2020). BLISS is a dialogue agent that tries to find what makes the people it converses with happy and healthy.

What is needed to do the research?

Data

Citation from research plan by Tom Sweers:

Since there is no Dutch negation corpus outside of the medical domain, an existing corpus needs to be annotated for negation.

Tools

An annotation environment is needed to annotate negation cues and scope. FLAT was chosen as a solution. This use case is covered by the FLAT and FoLiA support tasks.

Relevant issues that arose and were handled in the scope of this use case:

What software and services are involved?

  • FLAT
  • FoLiA

How to evaluate this?

References

  • Waterschoot, Jelte van et al. (May 2020). “BLISS: An Agent for Collecting Spoken Dialogue Data about Health and Well-being”. English. In: Proceedings of The 12th Language Resources and Evaluation Conference. Marseille, France: European Language Resources Association, pp. 449–458. isbn: 979-10-95546-34-4. url: https://www.aclweb.org/anthology/2020.lrec-1.57