Home // International Journal On Advances in Intelligent Systems, volume 5, numbers 3 and 4, 2012 // View article

Assessment Models and Qualitative and Symbolic Analysis Techniques for an Electrical Circuits eTutor

Authors:
Adrian Muscat
Jason Debono

Keywords: Electrical; Intelligent Tutoring System; Qualitative; Symbolic; Markov Model; Assessment;

Abstract:
This paper is about assessment models, domain expert models and user interfaces as components in an Intelligent Tutoring System that serves junior classes in electrical circuits. Two student models for the purpose of automated assessment are developed and tested. One of the models is a Markovian graph model, while the other is a histogram model. The effectiveness of these models in tracing the student's declarative as well as procedural knowledge is studied and compared to human assessment. The domain expert models are based on qualitative analysis and on symbolic quantitative techniques. These models are used to test declarative statements made by the student and also to generate a solution to the problem. The circuit analysis techniques are also studied from an educational point of view and are compared to numerical models on the basis of how much they help the student assimilate the knowledge. Two types of user interfaces are developed, one is text command line based, and the other comprises a graphical user interface. These three building blocks are used in the development of two independent systems, which are field tested with the engagement of polytechnic teachers and students at the higher national diploma level. The technical and pedagogical results obtained for the two modules are good and encouraging.

Pages: 278 to 290

Copyright: Copyright (c) to authors, 2012. Used with permission.

Publication date: December 31, 2012

Published in: journal

ISSN: 1942-2679