Additional Info

• Project title: Empathica – A Framework for Detecting Human Emotions from Decision-Making Patterns Using Machine Learning and Gamification
• Research project:

  Emotions play a central role in the day to day choices and decisions that people make. Automated emotion detection has therefore, become a widely used tool for a diversity of goals. However, there are scenarios when data for a person's facial image, speech in audio or transcribed forms may not be available and the data at hand is only for individual behavioural traits. We hypothesise that, emotional states can be inferred from behavioural patterns and to do it consistently with acceptable accuracy for a diverse population demographic, a computational model of human emotions for behavioural patterns is required, which to the best of our knowledge, does not exist at this time. This research will focus on addressing this gap by using machine learning techniques to create the said model. We will develop a framework based on conventional deep learning architectures in conjunction with gamification techniques for translating behavioural patterns into emotional states. This will be useful for inferring pre-determined factors of decision-making such as emotional and cognitive bias as well as intention, for instance, in automated user acceptance testing and detection of suicidal tendencies. We believe that the outcome of this research would enable us to usher a new dimension of interaction with computer systems, the 'Human-centric' computer interaction, where the system will be capable of knowing exact emotional state of its user and cater the user experience accordingly.

• E-mail:
• Primary Supervisor(s): Jesus Requena Carrion
• Secondary Supervisor(s): Riccardo Berta
• PhD cycle: XXXV