Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Beyond emotion archetypes: databases for emotion modelling using neural networks.

Roddy Cowie1, Ellen Douglas-Cowie, Cate Cox

  • 1School of Psychology, Queen's University Belfast. r.cowie@qub.ac.uk

Neural Networks : the Official Journal of the International Neural Network Society
|June 18, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A computerised test of perceptual ability for learning endoscopic and laparoscopic surgery and other image guided procedures: Score norms for PicSOr.

American journal of surgery·2017
Same author

The movements made by performers in a skilled quartet: a distinctive pattern, and the function that it serves.

Frontiers in psychology·2013
Same author

Politeness and social signals.

Cognitive processing·2011
Same author

Perceiving emotion: towards a realistic understanding of the task.

Philosophical transactions of the Royal Society of London. Series B, Biological sciences·2009
Same author

Can patients with chronic schizophrenia express emotion? A speech analysis.

Schizophrenia research·2003

Emotion research databases are evolving beyond traditional models to capture pervasive emotion in context. New approaches focus on naturalistic or induced, multimodal data for advanced applications like natural language processing.

Area of Science:

  • Affective computing
  • Human-computer interaction
  • Psychology

Background:

  • Traditional emotion research databases often rely on acted data and limited contexts.
  • The field requires more nuanced approaches to capture emotion as it occurs naturally in real-world interactions.
  • Existing database archetypes are becoming insufficient for current research needs.

Purpose of the Study:

  • To review the evolving landscape of emotion research databases.
  • To highlight the shift towards capturing pervasive emotion within diverse contexts.
  • To identify key features and emerging schemes for advanced emotion databases.

Main Methods:

  • Analysis of contemporary emotion research databases, focusing on their data collection strategies (naturalistic, induced).

Related Experiment Videos

  • Evaluation of database characteristics, including multimodality and contextual richness.
  • Review of emerging descriptive schemes for pervasive emotion.

    • Main Results:

    • The state of the art in emotion databases has moved beyond familiar archetypes.
    • There is a growing emphasis on capturing 'pervasive emotion' in action and interaction across various contexts.
    • Naturalistic and induced emotion data, alongside multimodal recordings, are increasingly important.
    • Applications extend beyond 'trouble shooting' to areas like 'really natural language processing'.

    Conclusions:

    • Emotion databases must accommodate pervasive emotion, context, and multimodality for advanced research.
    • Emerging descriptive schemes aim to capture the complexity of emotion expression.
    • The development of sophisticated, naturalistic emotion databases is crucial for future advancements in fields like NLP.