Joshua Poh-Onn Fan & Tina Kwai-Lan Mak
Department of Business Systems
University of Wollongong
joshua@uow.edu.au
A conventional approach to classroom student learning is through understanding the materials from the text book and then attempting the review problem exercises at the end of the chapters. This method only allows students to concentrate on the same set of problems. The advantage of this approach is that students can discuss and help each other but this often leads to plagiarism and students are not considering different forms of the same problem. The use of drilling in Computer Assisted Instruction (CAI) has resolved this problem partially. Unfortunely, CAI has deficiencies: CAI cannot provide reasoning and explanation for an answer provided by the student; CAI has difficulty in generating different forms of the same problem for a classroom environment where each student is not presented with the same problem; CAI can only do an exact match to the answer(s) expected. This paper describes a Knowledge-based Asisted Training and Testing System (KATTS) that has been developed to address some of the deficiencies in CAI. KATTS provides students with feedback by describing the line of reasoning used by the inference engine. KATTS allows the instructor to design problems by embedding variables in the problems and KATTS will use those different variables to generate different forms of the same problem. This feature provides greater flexibility in designing complex structural problems. KATTS can evaluate different answers to the different forms of the same problem by employing a knowledge-based induction technique. KATTS also provides management facilities for monitoring students performance and recording their scores. KATTS has been field tested in a second year accounting course in the last twelve months as a tool for self learning as well as for testing students under examination conditions. The field tests have been very sucessful and KATTS is starting to be used extensively in accounting courses at this University.
