9th Annual Conference of the International Speech Communication Association

Brisbane, Australia
September 22-26, 2008

An Interval Type-2 Fuzzy Logic System to Translate Between Emotion-Related Vocabularies

Abe Kazemzadeh, Sungbok Lee, Shrikanth S. Narayanan

University of Southern California, USA

This paper describes a novel experiment that demonstrates the feasiblity of a fuzzy logic (FL) representation of emotion-related words used to translate between different emotional vocabularies. Type-2 fuzzy sets were encoded using input from web-based surveys that prompted users with emotional words and asked them to enter an interval using a double slider. The similarity of the encoded fuzzy sets was computed and it was shown that a reliable mapping can be made between a large vocabulary of emotional words and a smaller vocabulary of words naming seven emotion categories. Though the mapping results are comparable to Euclidian distance in the valence/ activation/dominance space, the FL representation has several benefits that are discussed.

Full Paper

Bibliographic reference.  Kazemzadeh, Abe / Lee, Sungbok / Narayanan, Shrikanth S. (2008): "An interval type-2 fuzzy logic system to translate between emotion-related vocabularies", In INTERSPEECH-2008, 2747-2750.