An Enhanced Learner Evaluation Method Based on Psychophysiological Signals

Abstract

Learning is the process of acquiring new knowledge or revising existing knowledge. Among the several techniques to evaluate learners during a learning scenario such as; written or online exams, quizzes, spot–tests, etc, none of these techniques were capable enough to analyze the behavior of psychophysiological parameters during the learning scenario. This study suggests an approach to analyze the psychophysiological signals of a learner to evaluate the learner’s performance during the learning scenario. Supervised machine learning algorithms have utilized to identify whether the learner had previous knowledge about a particular learning content or not, by suggesting an approach to overcome certain limitations of the ordinary learner evaluating techniques. This study can be considered as an extension for an Intelligent Tutoring Systems (ITS) learner model, where the performance of the learners will be automatically evaluated without conducting the written/oral tests, exercises, quizzes, etc.