Fuzzy integral-based emotion inference on personalized V-A emotional space

Publication Date06 May 2014
DOIhttps://doi.org/10.1108/JSIT-11-2013-0057
Date06 May 2014
Pages138-148
AuthorIl-Kyoung Kwon,Sang-Yong Lee
SubjectInformation & knowledge management,Information systems
Fuzzy integral-based emotion
inference on personalized V-A
emotional space
Il-Kyoung Kwon
Department of Computer Science and Engineering,
Kongju National University, CheonAn-Si, Republic of Korea, and
Sang-Yong Lee
Division of Computer Science and Engineering,
Kongju National University, CheonAn-Si, Republic of Korea
Abstract
Purpose The focus in ICT development has shifted from performance to maximization of
tangibility. In particular, the interests in emotion-based services increase and more attention is paid to
improving the quality of service and maximizing tangibility. Hence, it is necessary to model human
emotions and utilize inferred emotions for services in order to enhance the quality of emotion-based
personalized services. To this end, this study aims to model personalized emotional space by means of
V-A emotion model and to suggest a fuzzy-based emotion inference method.
Design/methodology/approach – This study takes advantage of V-A emotion model, AHP and
fuzzy integral technique to implement an emotion inference based on the quantitative evaluation and
systematic expression of human emotions. By means of AHP, the personalized V-A emotion model
was established. Fuzzy measure and Choquet integral methods were applied for emotion inference
based on the inputs of valence and arousal (V-A) values, with the results evaluated in an experiment.
Findings – Thayer’s emotion model structures 12 human emotions in reference to V-A values. This
structure, however, may differ depending on individuals, and thus a method to measure and express it
appropriately is necessary. This study evaluates the relative weight of 12 emotions based on AHP and
regularizes them to compose the personalized V-A emotional space by drafting the V-A model. In
addition, fuzzy integral-based emotion inference has been tested in reference to the V-A values on the
established V-A emotion space.
Originality/value – Fuzzy integral-based emotion inference has been tested in reference to the V-A
values on the established personalized emotion space.
Keywords Emotion inference,Emotion modeling, Fuzzy integral,Fuzzy measure
Paper type Research paper
1. Introduction
As IT highly advances and becomes ever more complicated, the focus of ICT
development is changing from performance to maximization of convenience or
tangibility. In particular, the study on emotion-oriented services based on the subjective
dispositions of users has been expanded in the personalized service area. Accordingly,
it is necessary to model humanemotions appropriate for services and utilize the inferred
emotions in emotion-oriented personalizedservices. To this end, this study aimsto model
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1328-7265.htm
This research was supported by Basic Science Research Program through the National Research
Foundation of Korea (NRF) funded by the Ministry of Education (2012R1A1A2042625).
Received 8 November 2013
Revised 8 November 2013
Accepted 23 December 2013
Journal of Systems and Information
Technology
Vol. 16 No. 2, 2014
pp. 138-148
qEmerald Group Publishing Limited
1328-7265
DOI 10.1108/JSIT-11-2013-0057
JSIT
16,2
138

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT