Vocabulary Modeling of Social Emotion Based on Social Life Logging

摘要

Life logging in online space has recently been utilized to share emotion rather than to remember daily life.Therefore, this study is to determine social emotion in life logging. Social emotion was analized and collected from emotion vocabularies of 21 power blogging sites in Korea, such as Naver blog, Kakaotalk, Band, Facebook, Instagram, KaKaostory, and Daum. The collected vocabularies were sorted into morphemes and adjectives were extracted. The frequencies of the adjectives were analyzed, and their similarities was subjectively tested over 99 vocabularies showing high frequency. However, they showed insigniticant. In order to make up for the limition of this frequency analysis, 10 out of total 21 target sites were separately selected and 600 adjectives were extracted again. The extracted 600 vocabularies were sorted into 120 groups by Card sorting, and each group's representative vocabulary was defined as soial emotion demensions.