Combining the Overlay Model and Bayesian Networks to Determine Learning Styles in AHES

Combining the Overlay Model and Bayesian Networks to Determine Learning Styles in AHES

Copyright: © 2019 |Pages: 24
DOI: 10.4018/978-1-5225-5936-8.ch008
(Individual Chapters)
No Current Special Offers


First of all, it is important to note that the work presented here lies within the modeling part of the learner in an adaptive educational system construed as computational modeling of the learner. Modeling the learner in adaptive systems involves different information such as knowledge of the domain, the performance of the learning goals, background, learning styles, etc. Although there are several methods to manage the learner model, like the stereotype model or learner profiles, they do not handle the uncertainty in the dynamic modeling of the learner. The main hypothesis of this work is to show the link between the structure of the learner model and especially the characteristics of a learning profile and the learning style of a learning situation. This chapter shows how the combination of these two approaches to learner modeling can address the dynamic aspect of the problem in the modeling of the learner.
Chapter Preview


The technological landscape of the modern e-Learning is dominated by so-called learning management systems, learning management systems are powerful integrated systems that support a number of activities carried out by teachers and students during the e-Learning process. Teachers use e-Learning systems to develop course notes and quizzes on the Web, to communicate with students and to monitor and classify student progress. Students use it for learning, communication and collaboration.

Adaptive e-Learning systems often use learner models. A learner model is an internal representation of the user's properties, through which the system is based to adapt to the needs of each user. Before this model can be used, it must be built. This process requires a lot of effort to collect the required information and ultimately generate a learner model. Thus, an adaptive learning system takes all the properties of adaptive systems. To meet the needs of the application in the field of e-Learning, adaptive e-Learning systems adapt the learning material using user templates. The behavior of an adaptive system varies depending on the data from the learner's model and the learner's profile. Without knowing anything about the learner who uses the system, a system would behave in exactly the same way for all learners.

In general, the adaptation process can be described in three steps: retrieving user information, processing information to initialize the user model, updating the initialized user model and using the user's template to provide the adaptation. In the process of adaptation, it is possible to distinguish between two different characters; the learner or student with his goal of acquiring knowledge and the teacher. The goal of a teacher is to mediate the knowledge covered by a course to learners. Therefore, both points of view must be present in an e-Learning system.

Key Terms in this Chapter

Learning Situation: A part of a learning path that the learner takes to achieve a diploma or a certification.

Learning Path: A collection of learning situations that the learner takes in a certain period of time, it could be composed of a pretest, a learning activity or an evaluation.

E-Learning: A concept that describes the cognitive science principles of effective multimedia learning using electronic educational technology.

Learner Profile: A part of the learner model that only contain the static information of the learner that could be gathered before developing a learner model.

Adaptive Hypermedia Systems: An online information and help system, as well as institutional information systems that provide hyperlinks that are most relevant to the user in an effort to shape the user's cognitive load.

Bayesian Networks: Probabilistic graphical model or a type of statistical model that represents a set of random variables and their conditional dependencies via a directed acyclic graph.

Multi-Entity Bayesian Networks: A logic system that integrates first order logic (FOL) with Bayesian probability theory.

Complete Chapter List

Search this Book: