Abstract:

The aim of this paper is to present an adaptive learning management system to support an intelligent tutoring module. This system is responsible for the communication with the intelligent tutor and for the interaction with their end users through a web interface. This learning management system is part of the project ADAPT - Adaptive Learning Management System - conducted in collaboration between three institutions of higher education, Institute of Engineering - Polytechnic of Porto, Institute of Engineering - Polytechnic of Coimbra and the University of Trás-os-Montes and Alto Douro, and supported by Fundação para a Ciência e a Tecnologia from Portugal. ADAPT is capable of:
1) providing students with the learning object more suitable to their learning styles and preferences;
2) creating scripts (used to define learning paths) with adapted contents and learning styles;
3) recording the performance and the evolution in learning style of students as well as success and failure cases;
4) performing automatic learning based on success and failure, through case-based reasoning.

In the context of the ADAPT system, an adaptive script defines a sequence of nodes, representing contents and activities to be presented to the student. Each node contains alternative contents and activities suitable to different student profiles. These scripts allow representing alternative pedagogical models, in order to achieve a more interesting, varied, and productive learning. The selection of a script uses case-based reasoning. This mechanism considers the performance achieved previously by other students with similar profiles after using a set of available scripts.
The most predominant learning style is initially identified through a questionnaire. The evaluation of the learning style can evolve according to student‘s successes or failures in previous use of the system.

The intelligent tutoring module uses a combination of several models to define the student profile:
a) the Myer-Briggs Model to determine the personality profile;
b) the Kolb Model to describe the learning cycle by making a description of how an individual generates knowledge from its experience;
c) the Felder-Silverman Model to classify the learning style of a student;
d) the VARK (Visual, Aural, Read/write, Kinesthetic) model to determine student‘s learning preferences, i.e., the way in which a person has a greater willingness to receive the information.

The system also offers authoring tools to create learning objects, adaptive scripts and learning objects according to the Learning Object Metadata (LOM) standard. Currently, ADAPT is being tested in a class using the course of Digital Systems. In order to evaluate the benefits of the system, another class was used as control group where long-established learning methods were applied. Work is also being done with the aim of making this system adaptable to the user preferences, in addition to its adaptive capacities.

Acknowledgements: This work is supported by FEDER through COMPETE program and by National Funds through FCT under the project ADAPT (PTDC/CPE-CED/ 115175/2009).

Keywords:

Learning Management System, Adaptive Learning Management System, Intelligent Tutoring, Learning Style, Learning Object.