Re: [EMOXG] Updated discussion document, feedback requested

Hi,

we had a long and intense meeting today, and you will get the various 
pieces of the minutes in the next days. I think we made good progress.

Let me just answer this one point now, as it is scientifically 
interesting and quickly answered:

Idoia Zearreta schrieb:
> Core 7 - Agree with Ian again, we are not sure about a separate 
> intensity tag. Regarding with respect to dimensional classification, 
> isn't it enough using values of different scales? We think the arousal 
> scale is closely related to intensity. It might be useful for emotion 
> categories. Actually, we are not sure about this point.

I see two reasons for keeping intensity as a separate element, one 
scientific, the other one practical.

The scientific reason is that some models exist which associate a given 
position in three-dimensional PAD space (pleasure-arousal-dominance, 
from Mehrabian) with an intensity. The example I am thinking of right 
now is Patrick Gebhard's ALMA model: an emotion is triggered by a 
certain event; it is represented as a category but also has a (fixed) 
position in PAD space; and its intensity diminishes over time. He 
actually has a graphical display of this -- you see a red circle with 
diminishing radius at the given position of the PAD space. (it is freely 
available, see http://www.dfki.de/~gebhard/alma/)

So that means, intensity is not specific to categories.

The second reason is a practical one: By having intensity as a separate 
element, you can indicate a separate confidence for the intensity, 
independently of the category. The example I had given earlier is this one:

<emotion>
 <intensity value="0.1" confidence="0.8"/>
 <category set="everyday" name="boredom" confidence="0.1"/>
</emotion>

This is, we are highly confident that the emotion has a low intensity, 
but have only a vague guess at what emotion it could be. If intensity 
was an attribute, you would not be able to express this.

So, enough emotion markup for today ;-)

Best,
Marc


-- 
Dr. Marc Schröder, Senior Researcher at DFKI GmbH
Coordinator EU FP7 Project SEMAINE http://www.semaine-project.eu
Chair W3C Emotion ML Incubator http://www.w3.org/2005/Incubator/emotion
Portal Editor http://emotion-research.net
Team Leader DFKI Speech Group http://mary.dfki.de
Project Leader DFG project PAVOQUE http://mary.dfki.de/pavoque

Homepage: http://www.dfki.de/~schroed
Email: schroed@dfki.de
Phone: +49-681-302-5303
Postal address: DFKI GmbH, Campus D3_2, Stuhlsatzenhausweg 3, D-66123 
Saarbrücken, Germany
--
Official DFKI coordinates:
Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
Geschaeftsfuehrung:
Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313

Received on Friday, 24 October 2008 19:17:21 UTC