Emotion detection and its influence on popularity in a social network-based on the American TV series friends
Ilana Porter,Bar Galam,Roni Ramon-Gonen
DOI: https://doi.org/10.1007/s13278-023-01133-5
2023-09-26
Social Network Analysis and Mining
Abstract:4.59 billion people worldwide use social networks. Over a billion new posts are uploaded every day to Facebook's applications (Facebook, Instagram, WhatsApp). Social networks play a central role in consumption and marketing, politics, and social aspects. They affect users' decision-making processes, emotions, and behavior. Content on social media can include, for example, movies, pictures, and texts. The emotion expressed in the post content has an influence on the exposed individuals. Emotions can cause the user exposed to the content to follow another user, view more content raised by a certain user, share a post, or express support (or lack of support), thus affecting the popularity of the user. In this work, we examined the relationship between emotions expressed by a character and popularity measures. As a case study, we analyzed the texts presented in the popular series "Friends" over ten seasons. We found that women in the series express more emotions in general and, in particular, more emotions of anticipation, joy, trust, and fear. The findings show the relationship between different emotions expressed in the content and various popularity measures.