A Neural-Network system for Automatically Assessing Students
Duncan Mullier, David Moore, Leeds Metropolitan Univ., United Kingdom ; David Hobbs, Univ. of Bradford, United Kingdom
EdMedia + Innovate Learning, in Norfolk, VA USA ISBN 978-1-880094-42-6 Publisher: Association for the Advancement of Computing in Education (AACE), Waynesville, NC
This paper is concerned with an automated system for grading students into an ability level in response to their ability to complete tutorials. This is useful in that the student is more likely to improve their knowledge of a subject if they are presented with tutorial material at or just beyond their ability (Bergeron et al 1989). However, dynamically responding to a student's changing knowledge about a subject usually requires the presence of a human teacher, an altogether expensive resource. The system discussed here can grade both a student and the questions in a tutorial with minimal input from the human teacher. In order to accomplish this a specialist neural network is employed. The design and operation of our system is discussed along with arguments as to why a neural network approach is suitable for this problem.
Mullier, D., Moore, D. & Hobbs, D. (2001). A Neural-Network system for Automatically Assessing Students. In C. Montgomerie & J. Viteli (Eds.), Proceedings of ED-MEDIA 2001--World Conference on Educational Multimedia, Hypermedia & Telecommunications (pp. 1366-1371). Norfolk, VA USA: Association for the Advancement of Computing in Education (AACE).
© 2001 Association for the Advancement of Computing in Education (AACE)