Smart Assistant for Adaptive Course Preparation and Delivery in e-Learning Environments

Educated and skilled human resources and workers are real assets and a key of success and power for both nations and organizations. Therefore, education and training in general and Web-based Intelligent Tutoring Systems (ITS) in specific will expectedly play an important role in the future. Our first hypothesis in this research is that "adaptive" is a key for designing an effective ITS. Course adaptation must consider few essential dimensions: teaching strategies, student model that are based on background knowledge, learning style, and skills of each individual student, and the teaching approach suiting the instructor's cognitive model. According to their cognitive model, instructors, like all other experts in their fields, usually build on their selves' previous teaching experiences or the experiences of other experts. In this research, we investigated mechanisms supporting both authors and tutors in selecting the most appropriate learning materials for more effective learning outcomes. Authors need to prepare course materials that achieve specific objectives (e.g., syllabus and target skills). Students need to study with materials that match their learning styles and that build on their background knowledge. Therefore, our objective is to build a model and to suggest a framework and architecture for a Smart Instructor Apprentice (SIA) that provides instructors (authors and tutors) with intelligent Assistance in both Course Preparation and Delivery. SIA intelligently rewrites the course objectives according to educational theories and then adaptively selects the most appropriate Learning Objects (LO) from learning objects repositories (LORs) to align course objectives with students' models. SIA supports selecting the most appropriate LOs at both authoring and delivery stages of the educational process. SIA employs two main theories in building its model, namely, the revised Bloom's instructional design theory (RBT) and Felder & Silverman Learning style theory (FSLSM). In its model, SIA adapted the general structure of domain ontology to support RBT by accommodating new set of relations and a new definition of a concept. On the other hand, in order to support FSLSM, SIA suggested adding extra attributes to the LO's metadata attributes. SIA fulfils its job through a series of objectives rewriting steps and an LO selection strategy.