Emotions play an important role in cognitive processes, especially in learning tasks (Isen, 2000). However, in the context of Intelligent Tutors (ITS), the emotional factor has only recently been considered. However, modelling the emotional reactions of a learner during a learning session is an essential element for ITS to allow the mentor to predict when and how to intervene to help the learner perform his/her learning task under better conditions. In this article, we seek to predict the emotional reaction of the learner at a certain point in the learning. Our prediction approach is based on the causes that may have triggered this emotion and on its determinants such as personality. We therefore propose to solve this problem by using supervised machine learning algorithms and, more specifically, ranking algorithms.