Classification of human emotions from physiological signals using machine learning techniques

The project goal is to develop a system to classify the emotions of individuals with dementia. To achieve this goal, the applicant will have to design and test a machine learning algorithm that will extract important features from physiological signals such as heart-rate, skin conductance and skin temperature. Then, the algorithm will be trained using labels and insights generated by primary caregivers to automatically classify significant emotional states. Finally, the applicant will also work in the integration of the whole acquisition and processing systems through an app designed to work in any Android or iOS smartphone. This guarantees the versatility and portability of the system as a whole, and it will allow the users to use this system in any setting, where the caregivers will be able to identify, assess, and control the situations and places that cause distress in the users.

Faculty Supervisor:

Stefanie Blain-Moraes;Robert Kearney

Student:

Partner:

Aix-Marseille Université

Discipline:

Engineering

Sector:

Education

University:

McGill University

Program: