student profile: Mr Johan Alibasa


Thesis work

Thesis title: Multimodal Stress and Emotion Detection Using Physiological Signals and Behavioural Data trough Experience Sampling Method

Supervisors: Rafael CALVO , Abelardo PARDO

Thesis abstract:

Stress detection, or emotion detection generally, could give benefits to people by triggering intervention, e.g. emotion regulation. Emotion regulation has become an interesting topic especially since James Gross's review suggests that emotion regulation related research will give huge benefits, for instance, to decrease the impact of negative emotion by changing the way we think about the event and to help people preventing or postponing depression. Many studies have attempted to detect emotion and stress from physiological signal (from wearable device) and behavioural data (from smartphone) in real-world settings, but they have not considered the opportunity to implement emotion-triggered ESM which could produce high quality ground truth data for developing stress or emotion detection system. For detecting stress or negative emotions, behavioural data provided from activity tracking could inform the context as well as adding more features. However, it is still an open question whether these behavioural data could actually be used for input features of stress detection system. At the moment, we would like to assume that this additional modality, besides physiological data, could improve the system performance as several studies found that some activities (e.g. emailing and productive work) have high correlation with stress. To prove this hipothesis and build the model, we will collect both physiological and behavioural data from participants by running a study in real-world settings. In the end, the study will not only help to understand the relationship between emotion and wellbeing, but also to develop a sufficient emotion detection system which could be implemented for intervention system using mobile phone device.

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.