The Arousal video Game AnnotatIoN (AGAIN) Dataset is a huge corpus real-time arousal annotation traces of gameplay sessions. Each session is annotated by the players themselves, by watching the recorded footage right after they finish playing. The AGAIN dataset contains more than 1000 annotated gameplay sessions from nine games in three genres (platformers, shooters, and racing). The AGAIN dataset has already been used for a variety of machine learning tasks involving predicting arousal as well as for creating artificial agents that imitate the behavior or experience of AGAIN participants.
David Melhart, Antonios Liapis and Georgios N. Yannakakis: "The Arousal video Game AnnotatIoN (AGAIN) Dataset," in IEEE Transactions on Affective Computing 13(4), 2022. PDF BibTex
David Melhart, Antonios Liapis and Georgios N. Yannakakis: "Towards General Models of Player Experience: A Study Within Genres," in Proceedings of the IEEE Conference on Games, 2021. PDF BibTex
Matthew Barthet, Antonios Liapis and Georgios N. Yannakakis: "Go-Blend Behavior and Affect," in Proceedings of the ACII Workshop on What's Next in Affect Modeling?, 2021. PDF BibTex