No abstract

It's worth noting that the present paper lies within the range of modeling the learner in adaptive educational system as a conceptual modeling of the learner. Thought they are several methods that deal with the learner model; like stereotypes methods or learner profile…, but they are likely unable to handle the uncertainty embedded in the dynamic modeling of the learner. The present paper aims at studding different models and approaches to model the learner in an adaptive educational systems, and coming up with the most appropriate method based on the dynamic aspect of this model. The aim of this study is the argue that the learner model cannot be completely modeled based on one single method through the entire development process, but it needs a combination between several methods that will help for a complete modeling.

1.IntroductionThe user module is an essential component for Adaptive e-Learning systems. The term adaptation in e-learning systems involves the selection and the manner of presentation of each learning activity as a function focused on the entity having of knowledge and skills and other information given by each subject taught [1].Despite the different attempts to model the learning that is characterized by a dynamic aspect, we always find it difficult to achieve this goal [2]. The proposed approaches give us just a static view of the learner model, and don't manage the two domains of this model equally [3]. It doesn't cover the three stages of development of the learner model: the data collection, the initialization of the model and its update [4].In practice this model is in full development (the learner knowledge is evolving in the same module) [5].*Author for correspondence Thus, a dynamic view is essential. Therefore, in order to monitor the behavior of the learner in real time during a learning situation; we must adopt a model for dynamic management of the learning model, including the two types of domains of this model, and it three stages of initialization [6].The structure of this paper consists of combining Bayesian networks with stereotypes method to initialize the learner model in adaptive hypermedia educational systems (AHES). To achieve that goal, we will firstly return to the notions and definitions of the learner's model, the process of its development, its domains of information and their contents, and then we will concentrate on the Bayesian networks, their definitions and their rules of construction. Then, we will explain our approach for the initialization of the learner model based on a combination of the stereotype method and the Bayesian networks, precisely the field of intervention of each method and their expected results. Finally, and in order to disclose the validity of our hypothesis; we will present the experiments and the tests carried out. The results obtained from these experiments presented in this work are all arguments in favor of our

It's worth noting that the present paper lies within the range of modeling the learner in adaptive educational system as a conceptual modeling of the learner. The learner model is an essential component for Adaptive e-Learning systems. The term adaptation in e-Learning systems involves the selection and manner of presentation of each learning activity as a function that examines the entity of knowledge, skills and other information given by each subject taught. The present paper aims at studding the functionalities of the learner model in different Adaptive Hypermedia Educational Systems in the three stages of developing and managing this model. We will present in this comparative study, a full analysis of the learner model used in ten major hypermedias to come up with most appropriate method to treat the dynamic aspect of this model.

No abstract