Adaptive E-learning to Improve Communicative Skills of Learners with Autism Spectrum Disorder Using Eye Tracking and Machine Learning

Adaptive learning has proved its effectiveness in several research studies. It aims to offer a personalised and adaptive learning process to the needs and preferences of each learner using artificial intelligence technologies. Nevertheless, there is a category of learners who require special support in their learning, social life, nutrition and well-being. This category includes children with Autism Spectrum Disorder (ASD). The learning process for autistic children is based on a perception and support of the child's whole life. To this end, in this paper we aim to facilitate the social interaction of autistic learners through the development of their communicative skills. The approach is based primarily on the adoption of eye tracking technique to identify the emotions and interactions of autistic learners using a series of videos containing different emotional situations. Then, based on the results of the Naive Bayes classification algorithm, a series of learning social development activities will be presented to the learners using Applied Behaviour Analysis method (ABA). The results obtained will provide a more effective solution for helping learners to better integrate and interact with their social environment.