An Intentional Approach for Adaptive and Guided Construction of Educational Processes in Intelligent Tutoring System

Résumé

Our work is situated in the context of computer supported learning environments. It concerns more particularly the process adaptation in intelligent tutoring systems. In fact, the majority of these systems allow only the application of content adaptation and neglect the adaptation of educational processes (learning process and pedagogical process). This is a major constraint for providing personalized learning path and appropriate learning content and for exploiting the richness of individual differences of the learning needs and the pedagogical preferences. Within this context, the thesis seeks to answer the following central question: How to guide the educational processes construction according to relevant variables of educational context? To tackle this major problem, the key concept of our study is the modeling of educational process and its context. For that, the thesis proposes a new multi-levels view of educational process definition. This view is adopted also to study the process construction in different ITS by specifying a definition framework. By adopting this framework, a comparative study of different construction approaches identifies the major limits of an appropriate educational processes construction. To overcome these limits, we define a new approach to guide the adaptive construction of educational processes supported by an ITS. In fact, the proposed approach of construction must satisfy the following qualities: Generic: it is based on a generalized description of educational processes. Adaptive: it offers a dynamic adaptation of the learning path by considering the pedagogical preferences and the individual learning needs. Guided: this approach adopts a Map in order to guide the learning process construction by supporting the correlation between the learning process and the pedagogical process. Optimized: it specifies an algorithm that adopts flexible rules to optimize the construction of the learning process and the pedagogical process. The proposed approach includes the prediction of the educational context variables, the interpretation of predefined educational process models and the construction algorithm. The latter is based on different progression and consistency rules. The dynamic Bayesian network is used to predict the most relevant variables of educational context (the pedagogical orientation, the learning mode, the pedagogical method, the electronic media and the teaching style). These values are used to identify the appropriate intention by selecting the most suitable strategy. For this purpose, the approach specifies two predefined intentional models for educational process guidance, which are modeled by using Map formalism. In fact, these models couple the educational intentions with the educational strategies, and provide a multitude of paths between intentions.

Citer ce document

Accès au document

Ce document est hébergé sur son dépôt institutionnel d'origine.

Auteur(s)

Statistiques

Consultations : 2

Téléchargements : 0