Skip to Main Content
Todaypsilas vast use and frequent uploading of videos on the internet has created a need for a product that would search, select, and recommend videos that would be of interest to the user. Specifically, the video play list will be tailored to the userpsilas emotional mood. In this paper, we proposed XV-Pod, an emotion aware mobile video player that aims to tailor video selection based on a userpsilas emotional mood. In order to select videos that would be best suited to the user, the userpsilas emotional impact of videos was studied through physiological signals collected by BodyMedia SenseWear. We investigated two different approaches on recognizing userpsilas emotional response. In the first approach, we conducted an empirical study to find the emotional intensity change of the user before and after viewing the video. In the second approach, we tried to identify emotions of user by using a decision tree machine-learning algorithm. We discuss the results of both approaches. Finally, we discuss some implications of our findings and future work.