期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：2
出版社：Journal of Theoretical and Applied
摘要：It is necessary to find the human inter-rater agreement in emotion recognition research especially when handling with publicly available database. This paper discusses the Cohens Kappa coefficient technique to verify the actual tagged emotion categories for hybrid emotion model using music video as stimulus. This method has been done by finding the degree of inter-rater reliability between the five selected raters. As the results, the values of Cohens Kappa coefficients are over 0.87 for four actual tagged emotion categories which are happy, relaxed, sad and angry. These values demonstrate that the degree of inter-rater agreement are excellent. The actual tagged emotion categories are selected based on the division of average value of arousal-valence rating.
