Grants and Contributions:

Grant or Award spanning more than one fiscal year. (2017-2018 to 2022-2023)

Affective Computing is the associated field that focuses on the study and development of systems that can recognize, interpret, process and simulate human emotion. Medical biometrics denote physiological signals such as heart signatures, electrodermal responses of the skin and breathing rate, among others. The goal of the proposed work is to research methodologies for affective computing using medical biometrics with potential applications in healthcare for assessment of mental health, behavioral analysis in shopping and advertisement, detection of motivational characteristics in entertainment or education, among others. In the proposed work, human emotional states will be modeled, detected and classified using machine learning technology. Using information that results from biological responses, the most intimate expressions of emotions, the applicant and his team at the University of Toronto will research and develop algorithms that uncover hidden patterns of emotional behavior. For the first time, wearable technology will be used to measure bioresponses continuously in real-world settings. To date, the lack of a database of appropriate signals has limited any systematic development of emotion recognition technologies. Therefore, major objectives of the proposed research are: i) the investigation and development of pattern recognition algorithms suitable to characterize human emotions and be relevant to both military and non-military environments, and ii) the compilation of a suitable-scale database of emotionally-labeled physiological signals that is suitable for the evaluation of affective computing algorithms. The database will be made publicly available and it is expected that it will be endorsed by scientific teams performing research in the field of human emotion classification.

Despite the importance of the field, and in particular, its potential applications, emotion classification is still in its infancy and, to the best of the applicant’s knowledge, no comprehensive framework has been proposed for affective computing. It is expected that with the introduction of a much needed database of signals that meet the specifications of emotion classification tasks, the field of affective computing will fill a substantial gap and will be able to make steps towards a better and more systematic study and understanding of human emotions. The expected benefits of the proposed technologies on Canadian quality of life and economy are significant.