Emotion Recognition in E-learning Systems

Technological advances in e-learning systems offer new opportunities for students to reinforce academic development and to improve accessibility to education. Facial expression recognition is an increasingly important area in intelligent e-learning systems. However, supporting the emotional side of students during learning tasks is challenging and demands an awareness of students' emotions. For this purpose, a brief review is presented in order to underlie the importance role of facial expression recognition in e-learning systems. Its role in special education is investigated too. This paper centers on facial expression recognition using convolutional neural networks and its application in e-learning systems, by introducing a new system composed of three main steps: preprocessing, features extraction and classification. We tested the proposed system with students aged between 8 and 12 years old, in an educational game. The results showed that the proposed system reached state of the art results.

I. Introduction

Emotions have an important role on students learning and achievement. Emotions control the student’s attention, effect their motivation to learn and influence their self-regulation of learning. As stated by [1], self-regulated learning and motivation mediate the effects of emotions on academic achievement. Especially, positive emotions positively affect academic achievement when they are mediated by self-regulated learning and motivation.