Abstract-When leaning, be it a face to face or online, students favor a customized learning that meets their needs and preferences. Learners are more motivated and their evaluation results are satisfactory. Indeed, the adaptation of interventions according to the learning profiles of student is one of the best ways to improve learning. However, a learning profile is easier to detect in a face to face learning situation rather than in an online learning situation, especially when the defining rules the different profiles are imprecise and difficult to formulate in a digital language. In our contribution, we aim to solve this problem by proposing a profile deduction system allowing to translate the performance rules provided by the expert into numerical rules manipulated by the machine, which will facilitate the deduction of learning profiles from interactions made by learners face to training. For this, we will use the algorithm classification ANTClust. An experiment part is proposed to verify the accuracy of the classification performed and the obtained results.
Assistance systems require a growing interest from a large community of researchers especially in the context of online learning, their main purpose is to provide a proactive or reactive support to accomplish a learning activity. The diversity of roles and skills required in tutoring (perception of behaviour, interaction analysis, increasing motivation and regulation rhythm, etc.) makes the task of tutor arduous and difficult to manage. Thus, the assistance of tutors is the main focus of several research offering assistance of answers in the form of learning situations models and support tools for the various tasks that the tutor must ensure to his learners. However, elementary and repetitive tasks of tutor are rarely taken into account specifically for their proactive character. Therefore, we propose to support learning scenario of learner by an assistance scenario suitable for his learning style, on the basis of a set of pre-defined assistance objects. The assistance of tutors allows then, on the one hand, to advise them on interventions that best respond to the need of learner, and on the other hand, to trigger an automatic assistance according to specific conditions, discharging the tutor automatable repetitive and arduous tasks. Our contribution aims to relieve the tutor in his simple and monotonous tasks thanks to prediction method based on the Model of Assistance Object (MAO). The model aims to facilitate to tutor the definition of assistance objects more precisely. These assistance objects are used then by our prediction method to predict those most suitable depending on the learner's context, his learning style, and the feedback on the impact of objects of assistance which were previously presented.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.