Emerging opportunities for education in the time of COVID-19: Adaptive e-learning intelligent agent based on assessment of emotion and attention

The COVID-19 pandemic is a disastrous and rapidly evolving situation. It brings changes to many aspects of life, including teaching. A very important area for improvement of e- learning pedagogies is the ability to assess student engagement and the learning curve objectively and continuously. To this regard, we propose a novel procedure for personalized and adaptive assessment of learning performance based on methods for automated estimation of affective states. In this preliminary report we envision an intelligent agent which constantly monitors students’ behaviour parameters during online learning classes. Using unobtrusive video surveillance and machine learning the agent appraises key psychophysiological features related to emotion and attention.