eMotion: Estimation of User's Emotional State by Mouse Motions

Emotions play an important role in human interaction with humans and machines. Machines are not yet able to measure these and to interact with their users on an emotional level. The concept of affective computing seeks to change this as it would result in more enjoyable work with higher productivity and higher quality outcome. Different methods for emotion measurement have been developed, subjective and objective measures alike. This novel approach tries to connect closer to the user by using the direct interaction of the mouse that at the same time does not impede the user with its measurement mechanism. This work introduces the estimation of emotions by measuring minute motions in the user's mouse movements. It highlights the design and evaluation of a system that is capable of distinguishing user arousal levels and also provides an introduction to the field of affective computing.